cnhkmcp 2.1.3__py3-none-any.whl → 2.1.4__py3-none-any.whl

cnhkmcp/untracked/AI/321/206/320/261/320/234/321/211/320/255/320/262/321/206/320/237/320/242/321/204/342/225/227/342/225/242/knowledge/007_Finding_Consultant_Alphas_documentation.json

@@ -0,0 +1,333 @@
+{
+  "id": "getting-started-finding-consultant-alphas-read-first",
+  "title": "Finding Consultant Alphas",
+  "lastModified": "2025-11-18T10:21:25.763937-05:00",
+  "content": [
+    {
+      "type": "TEXT",
+      "value": "<p><b>Welcome to the WorldQuant BRAIN Research Consultant Community!</b></p><p>To kickstart your journey as a BRAIN Research Consultant, this article provides a step-by-step guide in navigating the new submission tests and platform features in the consultant environment, so that you can work towards finding your first consultant Alpha. At the start, the consultant environment may seem more challenging than the user environment you were used to, but treat this as the initial learning phase. Just like how you managed to find Alphas in the user environment, you’ll get the hang of the consultant environment eventually.</p>",
+      "id": "2e316b24-0e49-439b-b02b-d0ab27926326"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Understand the new submission tests"
+      },
+      "id": "bc9c1c14-9212-40da-898e-937fd1cb0076"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "finding_consultant_alphas_new_submission_tests.png",
+        "width": 858,
+        "height": 558,
+        "fileSize": 45873,
+        "url": "https://api.worldquantbrain.com/content/images/cmUHXK0-SgWtGXKF5gWMtAKzUQA=/440/original/finding_consultant_alphas_new_submission_tests.png"
+      },
+      "id": "1c98ab4b-bdd1-41e4-a149-7b30beb8cf4d"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p>Passing the new consultant submissions tests is one main source of challenge in the consultant environment. In addition to higher thresholds for passing the Sharpe and fitness tests, there will be two new submissions tests: Production Correlation and IS Ladder Sharpe.</p><p></p><p><b>Production Correlation</b></p><p>Your Alpha’s correlation will be tested against all existing Alphas in the consultant pool and not just your own. An Alpha passes the production correlation test if its maximum correlation with any BRAIN consultant Alpha is less than 0.7 or its Sharpe is at least 10% greater than the Sharpe of each Alpha with which it has a correlation greater than 0.7.</p><p></p><p><b><i>Tip:</i></b> By default, this test is not run immediately when you simulate an Alpha. When you test out a new Alpha idea, you may choose to run this test once first, by clicking the refresh button at the Production Correlation section in the Results panel. You can then assess whether this idea is unique enough from the rest of the consultant pool, before deciding whether to continue working on the idea. Try not to run this test too often though, as consultants may be limited to a certain number of correlation requests per hour.</p>",
+      "id": "548b418d-4a7b-42e7-9bfb-32496f41bd15"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "ConsultantProductionCorrelation.png",
+        "width": 624,
+        "height": 252,
+        "fileSize": 31813,
+        "url": "https://api.worldquantbrain.com/content/images/uiAg5kgIc9fgmDLlqGRhZEXbBug=/378/original/ConsultantProductionCorrelation.png"
+      },
+      "id": "3a6f30e9-e525-4f30-91ad-642d2bf85a59"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p><b>IS Ladder Sharpe</b></p><p>The IS Ladder Sharpe test checks whether your Alpha is robust throughout the simulation period, with more emphasis placed on recent years’ performance. It is an iterative test, where it assesses your Alpha’s Sharpe in the most recent 2 years in the first iteration, then the most recent 3 years in the next iteration, and so on, until we reach 10 years in the last iteration.</p><p></p><p>In each iteration, there are two benchmarks that the Alpha’s Sharpe will be compared against. The FAIL_THRESHOLD is the same as the usual overall Sharpe test you’re familiar with, which is 1.58 for Delay 1. The PASS_THRESHOLD is stricter and can differ for each year of iteration. For example, the PASS_THRESHOLD is 2.38 for years 2 to 5, 2.22 for year 6, 2.06 for year 7, etc.</p><p></p><p>In the first iteration, if the Alpha’s Sharpe for the most recent 2 years is below the FAIL_THRESHOLD of 1.58, the Alpha will fail the IS Ladder Sharpe test immediately. If it is above the relevant PASS_THRESHOLD of 2.38, the Alpha will pass the IS Ladder Sharpe test immediately. Otherwise, if it is above the FAIL_THRESHOLD but lower than the PASS_THRESHOLD, we will enter the next iteration where the Alpha’s Sharpe for the most recent 3 years will be compared to the two benchmarks.</p><p></p><p><b><i>Tip:</i></b> Some Alphas may have performed well in the past but deteriorated in recent years. Before you choose to double down on a particular Alpha idea, you may wish to check the Alpha’s performance in recent years. For example, is there a downward trend in its more recent PnL? Is the Alpha already failing the IS Ladder Sharpe test in the first iteration, which looks at the most recent two years?</p>",
+      "id": "e4e0e99b-621a-4a2c-bc2d-507e3a84dfd4"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "ISLadderSharpe.png",
+        "width": 624,
+        "height": 299,
+        "fileSize": 41806,
+        "url": "https://api.worldquantbrain.com/content/images/5u7pF9mnR2gDWsnTjEvmVe7yvqE=/379/original/ISLadderSharpe.png"
+      },
+      "id": "c2d2a2cb-9541-4d92-b58f-bc6dbb2ab1ea"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p><b>Other consultant submission tests</b></p><p>Here’s the full list of consultant submission tests and their details: <a href=\"https://platform.worldquantbrain.com/learn/documentation/consultant-information/consultant-submission-tests\">Consultant Submission Tests</a>.</p>",
+      "id": "78e6a7a7-7918-4d87-ad00-f5e1fd5aaf6d"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Start with what you’re familiar with"
+      },
+      "id": "20356008-2ee6-4835-b14b-b2ef7a86e457"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p>When you become a Research Consultant, you gain access to many new features: more datasets, a different simulation period, more regions, more simulation settings and so on. You may be wondering where to start from.</p><p></p><p>As consultants, your simulation period is now 10 years instead of 5 years. One way to begin would be to simulate your <b>past Alpha ideas</b> with this new simulation period, to see how they work over a longer period of historical data. You can also see how each Alpha performs in the new submission tests to earn a better understanding of how these tests work.</p><p></p><p>Thereafter, you can move on to other datasets in <b>dataset categories you’re more familiar with</b>. Swap out existing data fields you used in your past Alphas with similar data fields from these other datasets. Try new Alpha ideas using these datasets that you can understand well. You may also wish to explore <b>model datasets</b>, which usually has data fields with values that may already be treated and may be more easily used to create Alphas.</p><p></p><p>As you explore simulating Alphas in the consultant environment, you may wish to refer to tips on how to improve your Alphas here: <a href=\"https://support.worldquantbrain.com/hc/en-us/articles/10431497209623-A-list-of-must-read-posts-on-how-to-improve-your-Alphas-that-are-submitted\">A list of must-read posts on how to improve your Alphas that are submitted</a>.</p>",
+      "id": "97a7fb5f-a7bf-4ebf-8e1c-14f0db3fc972"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Try new regions"
+      },
+      "id": "6a8f6831-9516-44cd-9981-aa0869bd3f39"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "finding_consultant_alphas_region_characteristics.png",
+        "width": 2212,
+        "height": 320,
+        "fileSize": 32724,
+        "url": "https://api.worldquantbrain.com/content/images/8mOEo8heMpnd3c5LO1GbzJIJHkY=/445/original/finding_consultant_alphas_region_characteristics.png"
+      },
+      "id": "0b9abf3c-67ae-4da3-b4d0-cae6fbd32871"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p>When trying out Alpha ideas, you may choose to try the same idea out in all the available regions to potentially increase the likelihood of finding an Alpha that works. Here are some notes on the regions available on BRAIN.</p><p></p><p><b>USA</b></p><p>As users, you would have mostly worked with the USA region. This region remains popular among consultants and quite a number of Alphas have already been found in this region. This means that it may be hard to beat production correlation in USA.</p><p></p><p><b>HKG, KOR, TWN, CHN, IND</b></p><p>Look out for the high turnover in these regions. There may also be some region-specific considerations for KOR and TWN too. More tips here: <a href=\"https://platform.worldquantbrain.com/learn/documentation/advanced-topics/getting-started-hkgkortwn\">Getting started on HKG/KOR/TWN</a>, <a href=\"https://platform.worldquantbrain.com/learn/documentation/advanced-topics/china-region-consultants\">Getting Started: China Research for Consultants</a>, <a href=\"https://platform.worldquantbrain.com/learn/documentation/regions-and-universes/getting-started-india-alphas\"><b>India Alphas</b></a></p><p></p><p><b>GLB, EUR, ASI</b></p><p>For these regions, each region may contain stocks from different countries, which may cause the data field values to contain currency differences. It might be more meaningful to instead neutralize by country first before neutralizing by any other grouping. Refer to this guide on applying: <a href=\"https://platform.worldquantbrain.com/learn/documentation/advanced-topics/neut-users\">Double Neutralization</a>.</p>",
+      "id": "54f9b28d-f1b6-4d43-9a0f-17e820667ecd"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Explore new datasets"
+      },
+      "id": "1cb094ca-50f9-4c09-b667-e955a9c47d1c"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p>As you become more comfortable with the consultant environment, it may be time to try new dataset categories that you haven’t tried before! Trying new dataset categories can improve the diversity of your alpha pool too.</p><p></p><p>When trying new datasets, here are some things to take note:</p><ul><li>For vector data fields, you’ll need to convert the vector data field into a matrix data field first: <a href=\"https://platform.worldquantbrain.com/learn/documentation/advanced-topics/vector-datafields\">Vector Datafields</a>.</li><li>Evaluate new datasets using these tips: <a href=\"https://support.worldquantbrain.com/hc/en-us/community/posts/14833936207383--BRAIN-TIPS-6-ways-to-quickly-evaluate-a-new-dataset\">[BRAIN TIPS] 6 ways to quickly evaluate a new dataset</a>.</li><li>How to handle datafields with low coverage: <a href=\"https://support.worldquantbrain.com/hc/en-us/community/posts/14834608658199--BRAIN-TIPS-Weight-Coverage-common-issues-and-advice\">[BRAIN TIPS] Weight Coverage common issues and advice</a>.</li><li>Choosing datasets that are included in multiplier themes would increase your likelihood of having higher base payments: <a href=\"https://platform.worldquantbrain.com/learn/documentation/themes/overview-themes\">Overview of Themes</a>.</li></ul>",
+      "id": "2ef15b04-458e-4388-b256-54c1c2bf8019"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Build Alphas using research papers"
+      },
+      "id": "a527a3ad-051a-4baa-b963-3dfdf2e40b47"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p>The first and foremost step of building an alpha is coming up with an alpha idea. While there are multiple sources for finding ideas, research papers stand out as one of the most useful. In the beginning, the learn section of the platform is a good resource for sourcing ideas. For instance, the “Example” tab and Alpha concepts page are very useful. Although, for building low correlation alphas, it is important to use some unique or lesser utilized ideas. Research papers can assist you in finding different ideas.</p><p>Here are some general guidelines to build alphas using research papers:</p><p><b>Know your Operators</b>: The second step of the research process is to implement your alpha idea using operator. Hence, it is important to understand the concept of operators well. We recommend you go through the detailed operator descriptions before reading the research papers.</p><p><b>Reading the paper</b>: Visit any online journal and go through a few abstracts of the research papers. Choose a paper with an interesting idea to work with amongst all of the options. You can read the introduction, abstract, and main body. The aim is to gain a decent understanding of the research paper so that you know the methodology and data used by the authors.</p><p><b>Formulating your alpha</b>: The final step is to formulate your alpha. This is where your understanding of the operators comes in. Try to replicate the idea suggested in the paper and assess its performance, then think of any improvements that can be made on this version using any additional operators / data.</p><p>We will take you through an example <a href=\"https://papers.ssrn.com/sol3/papers.cfm?abstract_id=247435\">research paper</a> to show you how you can come up with an idea and refine it to create an alpha. This research paper talks about meeting or beating earnings estimate. Stocks that are better or equal to their earnings estimate are likely to show an outperformance as compared to other firms.</p><p><b>Iteration 1</b></p><p>Since the research paper is talking about an event (earnings release), we can use trade_when operator. A simple implementation of this idea could be to enter your positions when earnings estimate are beaten and amongst those stocks take positions according to the rank of ratio of actual earnings and estimated earnings and maintainyour position for the remainder of the quarter.</p><p>We have used the days_from_last_change operator to check if the earnings announcement was recent or not. The alpha below is in the US region, decay 4 and on subindustry neutralization, you can feel free to play around with the settings.</p><p></p>",
+      "id": "2ce7a039-f063-45f7-b4ae-10fbf79f5050"
+    },
+    {
+      "type": "SIMULATION_EXAMPLE",
+      "value": {
+        "settings": {
+          "instrumentType": "EQUITY",
+          "region": "USA",
+          "universe": "TOP3000",
+          "delay": 1,
+          "decay": 3,
+          "neutralization": "INDUSTRY",
+          "truncation": 5.0,
+          "pasteurization": "ON",
+          "unitHandling": "VERIFY",
+          "nanHandling": "OFF",
+          "language": "FASTEXPR",
+          "maxTrade": "OFF"
+        },
+        "type": "REGULAR",
+        "regular": "event=days_from_last_change(eps)<=5 && ts_delay(est_eps,10) <eps ;\r\ntrade_when(event,rank(eps/est_eps),-1)"
+      },
+      "id": "a7ee39b9-c9ee-4485-ae18-8a44fa64fa30"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>Here are the results</p>",
+      "id": "3599f67e-8669-4ff7-8891-19dd359e1e60"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "is_summary.png",
+        "width": 871,
+        "height": 109,
+        "fileSize": 16174,
+        "url": "https://api.worldquantbrain.com/content/images/W3F03AjtpctsJ2VwKlUHwEv9smQ=/170/original/is_summary.png"
+      },
+      "id": "378d2e62-06e3-4fbb-a819-455b7aaefcd4"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "Finding Consultant Alphas: Iteration 1 Graph",
+        "width": 735,
+        "height": 600,
+        "fileSize": 47533,
+        "url": "https://api.worldquantbrain.com/content/images/aEonuFZ7RGOSb7ZX1VdPlBnm87k=/138/original/Finding_Consultant_Alphas_Iteration_1_Graph.png"
+      },
+      "id": "e301a999-a0be-4d6d-bdf6-28d7e1dfffb9"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>The alpha isn’t submittable but the results aren’t too bad either. This is a good sign for us and increases our confidence in the idea since our first implementation has shown decent results. Performance metrics give us a direction that we need to focus on increasing the turnover.</p><p><b>Iteration 2</b></p><p>In our first iteration, we only focused on stocks that were meeting or beating the earnings estimate and hence we took positions in a small subset of stocks. We can try another implementation of our idea where we take positions in all stocks but take positive weight in those stocks that beat expectations.</p>",
+      "id": "b9cc136a-ddde-412d-bebf-f6c53703e0c0"
+    },
+    {
+      "type": "SIMULATION_EXAMPLE",
+      "value": {
+        "settings": {
+          "instrumentType": "EQUITY",
+          "region": "USA",
+          "universe": "TOP3000",
+          "delay": 1,
+          "decay": 3,
+          "neutralization": "INDUSTRY",
+          "truncation": 4.0,
+          "pasteurization": "ON",
+          "unitHandling": "VERIFY",
+          "nanHandling": "OFF",
+          "language": "FASTEXPR",
+          "maxTrade": "OFF"
+        },
+        "type": "REGULAR",
+        "regular": "event = days_from_last_change(eps)<=5 && ts_delay(est_eps,10) < eps ;\r\nif_else(event, 0.5 + rank(sales/assets), rank(sales/assets) )"
+      },
+      "id": "e950a720-63c6-4d06-9e78-97b507522dbc"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>The 0.5 + rank(sales/assets) ensures that we up weight stocks that meet or beat earnings estimate. The asset turnover ratio was used to get an idea of how efficiently the firm uses its resources, you can obviously try out different ratios or data fields as a proxy for financial health of the firm /as a default fundamental signal.</p>",
+      "id": "f26cc390-e708-41f1-8e42-a1905459b3ca"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "5.png",
+        "width": 879,
+        "height": 129,
+        "fileSize": 16335,
+        "url": "https://api.worldquantbrain.com/content/images/SJARGBfkiMIEV-3rAAcUGg9jyW8=/140/original/5.png"
+      },
+      "id": "7320d9da-0a5e-465c-bad8-9796c04cca4c"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "6.png",
+        "width": 888,
+        "height": 720,
+        "fileSize": 40737,
+        "url": "https://api.worldquantbrain.com/content/images/t7O6wXsjXZ9dPVCYedpjp6wSBvU=/141/original/6.png"
+      },
+      "id": "3c593653-0d99-4239-938a-50df30b1421e"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>Our turnover has increased since we have now moved from trade_when to if_else operator and take positions in all the stocks. We have taken a hit on our fitness due to this.</p><p><b>Iteration 3</b></p><p>Working on a new idea doesn’t mean you can’t use your previous knowledge or ideas to aid your research. In fact, a responsible use of your past knowledge with your new ideas can help you build a great number of alphas from a single research paper.</p><p>Our paper here talks about the outperformance of stocks that beat earnings estimate. So, deciding to go long on these stocks for the complete quarter sounds plausible. For the other stocks, it would make sense to use an idea which is irrespective of meeting or beating the earnings estimate. This will make sure we are not taking positions solely on the basis of information added at an interval of 3 months. From our past knowledge, we can bet on the mean reverting nature for these other stocks in our portfolio.</p>",
+      "id": "e4d7b579-aac7-4c81-889f-4eb85b9d8f2a"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "7.png",
+        "width": 578,
+        "height": 90,
+        "fileSize": 6855,
+        "url": "https://api.worldquantbrain.com/content/images/q1ksUMnw9AWVYr2Vs7K9yPDreng=/142/original/7.png"
+      },
+      "id": "b7686a65-9619-4bbc-b8a8-5923f584e4b3"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>Alpha =1 makes sure we are long on the stocks that beat earnings estimate since 1 is the max possible value of our alt_alpha hence even after neutralization it is likely that these stocks will have positive weight.</p>",
+      "id": "901f9bd8-94af-4167-90f6-5776f1ce5a4e"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "8.png",
+        "width": 887,
+        "height": 129,
+        "fileSize": 15781,
+        "url": "https://api.worldquantbrain.com/content/images/yImTywQUTKZB-f9uKjJBLzlJLMM=/143/original/8.png"
+      },
+      "id": "a9962409-0f3f-4d31-af83-932c46e34e02"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "9.png",
+        "width": 882,
+        "height": 721,
+        "fileSize": 33584,
+        "url": "https://api.worldquantbrain.com/content/images/50qIxwnV6awayEZ5TCtQyJQSzMw=/144/original/9.png"
+      },
+      "id": "11bf85e1-1dfa-4d8d-a300-c10f723f56a9"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>We see a significant improvement in all of our performance metrics. You may notice we have used the same event across all three implementations and despite that, achieved very different results. This shows how we can use a single framework and a responsible combination of different ideas to build multiple alphas.</p><p><b>Iteration 4</b></p><p>Similar to the previous iteration here we have scaled the alpha by sales_ps in order to adjust for different eps scales of different firms</p>",
+      "id": "6cad8874-9a99-426d-8fce-e8aeed1e6091"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "10.png",
+        "width": 526,
+        "height": 68,
+        "fileSize": 7455,
+        "url": "https://api.worldquantbrain.com/content/images/2ylOLtBHkKaROEG1zOjYz6GzZFM=/145/original/10.png"
+      },
+      "id": "fdfc8db5-78bd-40fc-8a96-b16c68369b4e"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "11.png",
+        "width": 881,
+        "height": 121,
+        "fileSize": 15902,
+        "url": "https://api.worldquantbrain.com/content/images/TI5j92ysrWCn8LuqBFBQTRU8jZw=/146/original/11.png"
+      },
+      "id": "a632bf6b-b49d-40df-b497-119ba3a347ea"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "12.png",
+        "width": 891,
+        "height": 725,
+        "fileSize": 32433,
+        "url": "https://api.worldquantbrain.com/content/images/GKdsncoy3eB6Zzh9_SEQJ_FODUk=/147/original/12.png"
+      },
+      "id": "679bef08-f3a6-4e3c-beb4-2cf00c46af23"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>You can use different implementations for the same idea on the brain platform or use any of the numerous data fields at your disposal to improve an idea or come up with an alpha that is not correlated to our existing production pool. So, reading new ideas/papers will not only help you build high-sharpe alphas but also in bringing down your correlation, which should be an equally important goal of your research. Further, the target is never to make an alpha “just good enough to be submittable” you may carry forward improvements even after passing all tests and submit the best version that you deem fit since the OS results hold the highest importance.</p><p><b>Some Additional Tips</b></p><p>Reading helps. Almost none of the ideas given in the research papers would work as they are presented (in some cases, opposite works). They are helpful in guiding you to think in a certain direction which you might otherwise miss completely.</p><p>Before reading, also make sure that the dataset given in the research paper exists in Brain. This will save you a lot of time in case you read a research paper only to realize the dataset isn’t available.</p>",
+      "id": "acea652f-aafb-4e67-948b-248b6938f425"
+    },
+    {
+      "type": "IMAGE",
+      "value": {
+        "title": "finding_consultant_alphas_best_practices.png",
+        "width": 1137,
+        "height": 598,
+        "fileSize": 37573,
+        "url": "https://api.worldquantbrain.com/content/images/wZCJpU91vCVM2xFw4wYgJqOTmgg=/442/original/finding_consultant_alphas_best_practices.png"
+      },
+      "id": "9b834c45-066b-4ff5-9df7-1df79b1d9939"
+    }
+  ],
+  "sequence": 3,
+  "category": "Consultant Information"
+}

cnhkmcp/untracked/AI/321/206/320/261/320/234/321/211/320/255/320/262/321/206/320/237/320/242/321/204/342/225/227/342/225/242/knowledge/007_Power_Pool_Alphas_documentation.json

@@ -0,0 +1,14 @@
+{
+  "id": "getting-started-power-pool-alphas",
+  "title": "Power Pool Alphas",
+  "lastModified": "2025-11-04T02:33:14.787086-05:00",
+  "content": [
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>Power Pool Alphas are simpler, smaller, higher quality Alphas that can help you do well across Genius, competitions and themes.</p><p></p><p><b>Eligibility Criteria for Power Pool Alphas Monthly Competition</b></p><p>Following are the list of criteria for an Alpha to be considered eligible for Power Pool Alphas:</p><ul><li>Sharpe &gt;= 1.0</li><li>Number of unique operators, including repeat operators &lt;= 8</li><li>Number of unique data fields (excluding grouping fields) &lt;= 3</li><li>grouping fields: country, industry, subindustry, currency, market, sector, exchange</li><li>Self-Correlation of just your Power Pool Alphas &lt; 0.5.</li><li>Once you tag an Alpha as Power Pool, it stays in the self-correlation pool even if you untag it later</li><li>Turnover should be between 1%-70% (both inclusive)</li><li>Pass Sub-universe check</li><li>USA D0 and D1, EUR D0 and D1, ASI D1, GLB D1 and OTHER ( JPN, CHN, HKG, TWN, KOR, AMR ) D0 and D1 research regions</li><li>Risk Handled Neutralization is mandatory for (USA/EUR/ASI/GLB/CHN Delay 1 and 0)</li></ul><p><b>Note:</b> If the Alpha has Self-Correlation among Power Pool Alphas &gt; 0.5, it should have Sharpe 10% higher than the most correlated Alpha to be considered eligible for submission.</p><p></p><p></p><p><b>Submission Checks Exempted for Power Pool Alphas</b></p><p>Following submission checks do not apply for Power Pool Alphas:</p><ul><li>Prod Correlation</li><li>Self Correlation with own Alphas from outside Power Pool</li><li>Fitness threshold</li><li>IS Ladder</li></ul><p></p><p><b>Monthly Awards Qualification</b></p><ul><li>To qualify for the monthly awards in any of the 8 <a href=\"https://platform.worldquantbrain.com/competition/consultant\">Power Pool leaderboards</a>, a minimum of 10 tagged Alphas per leaderboard is required, with no minimum submissions required per month​. Those 10 Alphas could have been submitted even prior to the month</li></ul><p></p><p><b>Submission Quotas After 3 Months</b></p><p>The below quota will apply when consultants cross 3 months since date of first submitting a Power Pool Alpha:</p><ul><li>Each consultant can submit up to 10 pure Power Pool Alphas per calendar month, in one single leaderboard of Power Pool​. Pure Power Pool Alphas are those that do not meet submission criteria of either Atom or Regular Alphas</li><li>Each consultant can submit up to 1 pure Power Pool Alpha per day</li><li>Alpha classified as [Power Pool + ATOM] or [Power Pool + Regular] is excluded from these two limits</li><li>Example of daily submissions:<ul><li><ul><li>1 Pure Power Pool Alpha</li><li>1 [Power Pool + ATOM] Alpha</li><li>Total: 2 Power Pool-related submissions</li></ul></li></ul></li><li>For consultants who have not crossed 3 months since their first Power Pool Alpha submission, standard submission limits for BRAIN consultants apply. Max 4 alpha submissions in a day.</li></ul><p></p><p><b>Description of the Power Pool Alpha</b></p><p>To submit a Power Pool Alpha, it is mandatory to describe the idea in at least 100 characters. In the PROPERTIES section at the bottom of the Simulation results. Using the template of Idea and Rationale. Otherwise the Alpha will not be eligible for Power Pool.</p><p>Here is an example description:</p><ul><li><b>Idea</b>: In normal market conditions, if a stock is shorted more, its likelihood of bouncing back may also increase (reversion). However, in extreme cases where the consensus in the market is high reflecting in extremely high/low level of short interest, it may potentially be better to follow that trend</li><li><b>Rationale for data used:</b> shrt3_bar field is a vector data field representing the demand to borrow stock, with higher values indicating higher demand</li><li><b>Rationale for operators used:</b></li><li>vec_avg(): Calculates the average value of shrt3_bar for a given day</li><li>Conditional operator: Separates normal cases from extreme ones</li><li>ts_backfill: Handles NaN values in the data field, detected by checking the coverage with a visualization tool</li></ul><p><b>How do I add or remove Alphas from the Power Pool?</b></p><p>You can view the list of your Power Pool Alphas on the Alphas Page under the <a href=\"https://platform.qa.worldquantbrain.com/alphas/submitted\">Submitted</a> tab. To remove a submitted Alpha from the Power Pool, go to the Submitted tab, open the Alpha description, and click the cross next to the \"PowerPoolSelected\" tag in the Properties section.</p><p><b>Please note that even after removing the tag, this Alpha will still be part of the self-correlation pool, so new Power Pool Alphas will still be checked for correlation against it.</b></p><p>To retag a submitted Alpha later to the Power Pool, in the Properties section of the Alpha, click on the Tags dropdown and retag the alpha as \"PowerPoolSelected\"</p><p></p><p><b>Tips to create Power Pool Alphas</b></p><ul><li>Review Merged Performance before tagging the Alpha</li><li>Explore Low turnover Alphas and liquid (small) universes, which can help improve <a href=\"https://support.worldquantbrain.com/hc/en-us/community/posts/29647491881623-How-to-Improve-After-Cost-Performance\">After Cost performance</a></li><li>Explore your existing Alpha pool for eligible signals</li><li>Make use of the newly released <a href=\"https://support.worldquantbrain.com/hc/en-us/articles/30816357468183-How-can-I-make-use-of-Investability-Constrained-PnL-for-excelling-in-Power-Pool-Alphas-Competition\">Investability Constrained PNL</a></li><li>Your pool should have diverse signals to ensure robust performance of your pool in OS. Diversify your pool across datasets, ideas, operators, universes and even turnover (to some extent)</li><li>After creating a sizable pool of Power Pool Alphas, consider removing Alphas which have high correlation with other Alphas in your Power Pool. This may potentially improve combo performance while reducing the Alpha count penalty</li><li>Reach out to your research advisor for specific tips on creating good Power Pool Alphas</li></ul>",
+      "id": "e19d6aea-bc96-43be-bd68-9083bb48df15"
+    }
+  ],
+  "sequence": 2,
+  "category": "Consultant Information"
+}

cnhkmcp/untracked/AI/321/206/320/261/320/234/321/211/320/255/320/262/321/206/320/237/320/242/321/204/342/225/227/342/225/242/knowledge/007_Research_Advisory_Program_documentation.json

@@ -0,0 +1,35 @@
+{
+  "id": "brains-new-advisory-program",
+  "title": "Research Advisory Program",
+  "lastModified": "2024-10-04T15:32:18.609549-04:00",
+  "content": [
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "The BRAIN Research Advisory Program"
+      },
+      "id": "3a3a103c-4d3a-4fbb-b429-2af54901ae26"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p>The BRAIN Research Advisory Program is available for all BRAIN Research Consultants to join and take advantage of direct access to a WorldQuant Researcher.</p><p><b>What:</b> You will be given the opportunity to speak with and learn from one of our WorldQuant Researchers on key areas of alpha research specific to your assigned Researcher’s specialty. This Advisory Program is an <b>optional resource</b> for each of you to continue your quant journey on BRAIN and hopefully gain deeper insight in your research methodology and process.</p><p><b>Key topics:</b> When communicating with your assigned Researcher, You can focus on many different topics, including:</p><ul><li>Alphas research in much greater depth (related to the areas of focus of your assigned Researcher);</li><li>SuperAlpha research;</li><li>Personalized advice on refining your ideas;</li></ul><p><b>How often:</b> The BRAIN Researcher will schedule a monthly session for this group for you to join as you wish. This may take place either in-person or may be conducted virtually. While we expect these sessions will be helpful to all BRAIN Consultants, there is no requirement or obligation to join the sessions. They are meant solely for your benefit.</p><p><b>Why:</b> We believe that this will be a unique experience enabling you to be part of a cohort of fellow learners. These sessions will be structured with concrete learning objectives, and you will have access to continuous engagement throughout the year with your assigned Researcher. You will be given the opportunity to receive first-hand insight into how your assigned Researcher tackles a problem or comes up with a particular idea or methodology.</p><p></p><p></p>",
+      "id": "2b076535-150b-4df0-957e-e024c1d26dc7"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Find Your Advisor"
+      },
+      "id": "ca6f366d-233d-4bdc-927f-85fc52dc2a41"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p>You can find out more on your advisor and the content recommended by them on the next page.</p>",
+      "id": "296d959d-470c-4530-a2be-f61f1020026d"
+    }
+  ],
+  "sequence": 100,
+  "category": "Consultant Information"
+}

cnhkmcp/untracked/AI/321/206/320/261/320/234/321/211/320/255/320/262/321/206/320/237/320/242/321/204/342/225/227/342/225/242/knowledge/007_Starting_Guide_for_Research_Consultants_documentation.json

@@ -0,0 +1,14 @@
+{
+  "id": "get-started",
+  "title": "Starting Guide for Research Consultants",
+  "lastModified": "2025-11-04T02:38:47.552711-05:00",
+  "content": [
+    {
+      "type": "TEXT",
+      "value": "<p></p><p><b>Compensation and Additional Opportunities</b></p><p>As a WorldQuant Brain Research Consultant, you are eligible to receive compensation in the form of periodic <a href=\"https://support.worldquantbrain.com/hc/en-us/articles/4672184130455-How-can-you-earn-more-Base-payment-\">base payments</a>, as well as <a href=\"https://support.worldquantbrain.com/hc/en-us/articles/4702810979351-What-is-the-Quarterly-Payment-\">quarterly payments</a> based on the quality of your alphas. In each case, please refer to your individual consulting or service agreement with WorldQuant for details on such compensation.</p><p>In addition, you may be eligible to receive a referral fee from time to time based on successful referrals under the <a href=\"https://platform.worldquantbrain.com/referral\">referral program</a>.. You may also be considered for full-time positions and internship opportunities at WorldQuant and potentially invited to Brain’s exclusive knowledge sharing and networking meetups from time to time.</p><p><b>New Submission Tests</b></p><p>As a Research Consultant, your alphas will go through a stricter set of tests in order to clear the <a href=\"https://platform.worldquantbrain.com/learn/documentation/consultant-information/consultant-submission-tests\">submission criteria</a>. For instance, each of your alpha’s correlation will be tested against all alphas in the consultant pool and not just your own.</p><p><b>Additional simulation results</b></p><p>As mentioned in the previous paragraph, one of the new features will provide you access to a <a href=\"https://platform.worldquantbrain.com/learn/documentation/consultant-information/visualization-tool\">visualization tool</a> as part of your simulation results. You can check the simulated (i) coverage, (ii) PnL, (iii) Size and (iv) Sharpe of your alphas across various sectors and industries. This will help you identify whether or not your alpha has a concentration bias for any particular industry or sector and may provide insights on any potential improvements to your alpha.</p><p><b>More datasets and regions</b></p><p>You will also have access to the European, Chinese and Asian markets / universes after becoming a research consultant. In addition to these regions, you will also have access to <a href=\"https://platform.worldquantbrain.com/data/data-sets\">consultant only datasets</a>. New datasets help you in implementing various new alpha ideas as well as changing the implementation of your existing alpha ideas. A good way to start with new datasets could be implementing your existing alpha ideas with these datasets so that you can build an intuition for the same. All datasets have their descriptions listed alongside their type (vector/matrix) and their coverage across different regions. Combining base data fields is a good idea but remember to avoid overfitting by sticking to your original alpha idea.</p>",
+      "id": "2ea32aa5-daef-4857-8fb1-8ab6839d5e8f"
+    }
+  ],
+  "sequence": 1,
+  "category": "Consultant Information"
+}

cnhkmcp/untracked/AI/321/206/320/261/320/234/321/211/320/255/320/262/321/206/320/237/320/242/321/204/342/225/227/342/225/242/knowledge/007_Visualization_Tool_documentation.json

@@ -0,0 +1,99 @@
+{
+  "id": "visualization-tool",
+  "title": "Visualization Tool",
+  "lastModified": "2024-03-24T16:45:39.384441-04:00",
+  "content": [
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Visualization Tool"
+      },
+      "id": "9347f953-64b8-475b-a2e9-2fdde680d8bd"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p>The visualization feature allows users to plot graphs of Alpha values and statistics, in addition to the PnL graph. It enables users to analyze the visualized output and fine tune Alphas before re-simulating or submitting.</p><p>Running the visualization has an impact on simulation speed, so be judicious in utilizing this feature. By default this feature is turned off but can be turned on in simulation settings. Visualization data for older Alphas is only available for a week after simulation date.</p><p></p>",
+      "id": "352f243d-7975-42b5-85cb-6f355601e702"
+    },
+    {
+      "type": "TABLE",
+      "value": {
+        "data": [
+          [
+            "Term",
+            "Usage"
+          ],
+          [
+            "Coverage",
+            "Number or percentage of instruments with non-NAN Alpha values"
+          ],
+          [
+            "Size",
+            "Amount of money allocated as percentage of booksize"
+          ],
+          [
+            "PNL",
+            "PNL attributed to the relevant set of instruments"
+          ],
+          [
+            "Sharpe",
+            "Sharpe value of associated PNL stream"
+          ],
+          [
+            "Capitalization",
+            "Quintile of market capitalization of underlying stock"
+          ],
+          [
+            "Industry / Sector",
+            "Market grouping (scroll down to the bottom of the page for the full list of sectors and industries)"
+          ]
+        ],
+        "firstRowIsTableHeader": true,
+        "firstColIsHeader": false
+      },
+      "id": "b74b4c0f-a7e6-41ab-88fc-8cb3d95c2968"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Available Graphs by Category"
+      },
+      "id": "919640f3-5820-4aa9-a527-ddc15a5ee931"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p><b>Coverage</b></p><ul><li>Alpha Coverage</li><li>Coverage by Industry</li><li>Coverage by Sector</li></ul><p><i>‘Coverage’ denotes the average number of stocks whose value change each day.</i></p><p><b>Size</b></p><ul><li>Average Size by Capitalization</li><li>Average Size by Industry</li><li>Average Size by Sector</li></ul><p><b>PNL</b></p><ul><li>PNL by Capitalization</li><li>PNL by Industry</li><li>PNL by Sector</li></ul><p><b>Sharpe</b></p><ul><li>Sharpe by Capitalization</li><li>Sharpe by Industry</li><li>Sharpe by Sector</li></ul><p><b>Others</b></p><ul><li>Turnover with Time</li><li>Industry Average value with Time</li><li>Sector Average value with Time<br/></li></ul>",
+      "id": "696c41ee-ca6c-4d9f-ba0e-c3e868ddc766"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Tips to Analyze the Graphs"
+      },
+      "id": "6cf17836-aa42-4036-833c-de3c5d529ad0"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><ol><li>Highly varying turnover or average Alpha value with time should be avoided.</li><li>Very large booksize concentrated on low cap or one sector or industry is not good, as it potentially entails low liquidity or diversification.</li><li>Filtering out a lot of stocks in your simulation leading to less diversification across caps, sector or industry is not a good practice. This could be checked using the coverage plots.</li><li>Sharpe contribution from different industries and sectors towards the overall Sharpe indicates diversification.</li><li>Good Sharpe in high cap stocks along with low caps makes a good Alpha.</li></ol>",
+      "id": "92a48b09-d453-4023-b861-da530967eb96"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Full List of Sectors & Industries"
+      },
+      "id": "06528010-b659-44eb-afbd-549a0f6772a2"
+    },
+    {
+      "type": "TEXT",
+      "value": "<p></p><p><b>Sector</b></p><ul><li>Distribution Services</li><li>Communications</li><li>Transportation</li><li>Utilities</li><li>Retail Trade</li><li>Miscellaneous</li><li>Consumer Durables</li><li>Industrial Services</li><li>Commercial Services</li><li>Consumer Non-Durables</li><li>Process Industries</li><li>Consumer Services</li><li>Technology Services</li><li>Producer Manufacturing</li><li>Health Technology</li><li>Energy Minerals</li><li>Electronic Technology</li><li>Non-Energy Minerals</li><li>Finance</li><li>-99999 (Uncategorized)</li></ul><p><b>Industry</b></p><ul><li>All Unclassified Establishments</li><li>Public Administration</li><li>Other Services (except Public Administration)</li><li>Educational Services</li><li>Arts Entertainment and Recreation</li><li>Agriculture Forestry Fishing and Hunting</li><li>Health Care and Social Assistance</li><li>Accommodation and Food Services</li><li>Administrative and Support and Waste Management and Remediation Services</li><li>Transportation and Warehousing</li><li>Construction</li><li>Wholesale Trade</li><li>Utilities</li><li>Retail Trade</li><li>Real Estate and Rental and Leasing</li><li>Management of Companies and Enterprises</li><li>Professional Scientific and Technical Services</li><li>Information</li><li>Finance and Insurance</li><li>Mining Quarrying and Oil and Gas Extraction</li><li>Manufacturing</li><li>-99999 (Uncategorized)</li></ul>",
+      "id": "c653be59-a1e9-4fef-8537-835618e97cf1"
+    }
+  ],
+  "sequence": 11,
+  "category": "Consultant Information"
+}

cnhkmcp/untracked/AI/321/206/320/261/320/234/321/211/320/255/320/262/321/206/320/237/320/242/321/204/342/225/227/342/225/242/knowledge/007_Your_Advisor_-_Kunqi_Jiang_documentation.json

@@ -0,0 +1,53 @@
+{
+  "id": "your-advisor-kunqi-jiang",
+  "title": "Your Advisor - Kunqi Jiang",
+  "lastModified": "2024-12-04T06:02:14.075394-05:00",
+  "content": [
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Advisor: Kunqi Jiang"
+      },
+      "id": "b60e9b6d-fc1d-4da2-9c07-f092ce55e19b"
+    },
+    {
+      "type": "TEXT",
+      "value": "<h3><a href=\"https://linkedin.com/in/kunqijiang\">LinkedIn Profile</a></h3>",
+      "id": "8b1226ce-3e14-46d6-b9f8-d464abe8ce67"
+    },
+    {
+      "type": "TEXT",
+      "value": "<ul><li>I am excited to introduce myself and collaborate with you during this BRAIN Research Advisory Program.</li><li>I earned my Master degree in Machine Learning from University College London in 2018. After that I worked for several internet company such as NetEase and Baidu as an engineer in recommendation and search engine algorithm.</li><li>I joined Brain as a researcher after won 3rd Place in national stage of Global Alphathon 2022 where I can unleash my interest in Quantitative finance for many years.</li><li>My research focus is machine framework that can be flexibly incorporate human knowledge to create alphas with logic and financial sense.</li><li>I am looking forward to meeting you soon to discuss topics that you are interested to learn and grow through the BRAIN Research Advisory Program.</li></ul>",
+      "id": "6b0e39d0-a793-482a-b25e-3db783c86ddc"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Contacting your advisor"
+      },
+      "id": "a61850f7-fa12-422a-9ea5-9635d3830818"
+    },
+    {
+      "type": "TEXT",
+      "value": "<ul><li>点此预约您的Advisor会议 <a href=\"https://calendly.com/mainlandchina/1-1-meeting-with-brain-cn-hk-researcher-kunqi-jiang\">https://calendly.com/mainlandchina/1-1-meeting-with-brain-cn-hk-researcher-kunqi-jiang</a> </li><li>Register for the next session from the <a href=\"https://platform.worldquantbrain.com/events\">Events Page</a></li><li>Advisory forum:<ul><li><a href=\"https://support.worldquantbrain.com/hc/en-us/community/topics/12913416465431-%E4%B8%AD%E6%96%87%E8%AE%BA%E5%9D%9B\">中文论坛 – WorldQuant BRAIN</a></li><li><a href=\"https://support.worldquantbrain.com/hc/en-us/community/topics/18910956638743-%E9%A1%BE%E9%97%AE%E4%B8%93%E5%B1%9E%E4%B8%AD%E6%96%87%E8%AE%BA%E5%9D%9B\">顾问专属中文论坛 – WorldQuant BRAIN</a></li></ul></li></ul>",
+      "id": "7e65c27b-7cdf-492c-9e3f-02a6d60601a2"
+    },
+    {
+      "type": "HEADING",
+      "value": {
+        "level": "1",
+        "content": "Curated content from the Advisor"
+      },
+      "id": "6d4871da-5df5-4db8-86cc-051612e1b4da"
+    },
+    {
+      "type": "TEXT",
+      "value": "<ul><li> <a href=\"https://support.worldquantbrain.com/hc/en-us/community/posts/27928616328855--%E6%96%B0%E4%BA%BA%E5%BF%85%E8%AF%BB-%E6%96%B0%E9%A1%BE%E9%97%AE%E5%85%A5%E9%97%A8%E6%8C%87%E5%8D%97-%E5%86%85%E5%90%ABBRAIN-API%E5%92%8C-%E9%A1%BE%E9%97%AE%E6%94%B6%E5%85%A5%E7%AD%89%E7%B2%BE%E5%8D%8E%E5%B8%96\">【新人必读】新顾问入门指南！(内含BRAIN API和 顾问收入等精华帖） – WorldQuant BRAIN</a> </li><li> <a href=\"https://support.worldquantbrain.com/hc/en-us/community/posts/27928747887383--%E8%BF%9B%E9%98%B6%E5%AD%A6%E4%B9%A0-Human-Alpha%E4%BB%8E%E5%85%A5%E9%97%A8%E5%88%B0%E7%B2%BE%E9%80%9A-%E5%86%85%E5%90%AB100-Alpha-Idea%E5%88%86%E4%BA%AB%E4%B8%8E%E6%8E%A8%E8%8D%90%E8%AE%BA%E6%96%87\">【进阶学习】Human Alpha从入门到精通（内含100+ Alpha Idea分享与推荐论文） – WorldQuant BRAIN</a> </li><li> <a href=\"https://support.worldquantbrain.com/hc/en-us/community/posts/27928714517015--%E8%BF%9B%E9%98%B6%E5%AD%A6%E4%B9%A0-Machine-Alpha-%E4%BB%8E%E5%85%A5%E9%97%A8%E5%88%B0%E7%B2%BE%E9%80%9A-%E5%86%85%E5%90%AB%E6%8E%A8%E8%8D%90%E7%9A%84%E6%95%B0%E6%8D%AE%E4%B8%8E%E6%A8%A1%E6%9D%BF%E5%90%88%E9%9B%86\">【进阶学习】Machine Alpha 从入门到精通（内含推荐的数据与模板合集） – WorldQuant BRAIN</a> </li><li> <a href=\"https://support.worldquantbrain.com/hc/en-us/community/posts/27293949853591--%E5%B9%B3%E5%8F%B0%E4%B8%8A%E6%96%B0-%E9%99%90%E6%97%B6%E7%BD%AE%E9%A1%B6-%E4%B8%80%E6%96%87%E5%B8%A6%E4%BD%A0%E8%AF%BB%E6%87%82-Genius-Program-%E9%AB%98%E9%98%B6%E9%A1%BE%E9%97%AEQuarterly-Payment%E6%9C%89%E6%9C%BA%E4%BC%9A%E8%8E%B7%E5%BE%97%E9%AB%98%E8%BE%BE8000USD%E4%BF%9D%E5%BA%95\">【平台上新，限时置顶】一文带你读懂 “Genius Program\" -- 高阶顾问Quarterly Payment有机会获得高达8000USD保底 – WorldQuant BRAIN</a> </li></ul>",
+      "id": "e1401246-6997-4702-a1df-82e1da0a7549"
+    }
+  ],
+  "sequence": 107,
+  "category": "Consultant Info"
+}