{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":790393788,"defaultBranch":"main","name":"Portfolio-Mgmt-And-Machine-Learning-in-Finance","ownerLogin":"Agoons20","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2024-04-22T19:52:03.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/56331600?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1713815524.649128","currentOid":""},"activityList":{"items":[{"before":"15fe2513054e1fd3087b1d6419fc1437f730a58f","after":"efeb83d151932239103e582b7323172253c7e3e1","ref":"refs/heads/main","pushedAt":"2024-05-09T06:54:06.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Rename README.md to README.md \n\nOkay","shortMessageHtmlLink":"Rename README.md to README.md"}},{"before":"33d91d42fae3fcae16fd0c34be96bb1c4c6e3242","after":"15fe2513054e1fd3087b1d6419fc1437f730a58f","ref":"refs/heads/main","pushedAt":"2024-05-09T06:53:20.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Update README.md\n\nokay","shortMessageHtmlLink":"Update README.md"}},{"before":"f436aa1e7adfbe99422d03f9a398cd6b0de8c56a","after":"33d91d42fae3fcae16fd0c34be96bb1c4c6e3242","ref":"refs/heads/main","pushedAt":"2024-05-09T06:26:02.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Update README.md","shortMessageHtmlLink":"Update README.md"}},{"before":"feef420eee099969bc83dbb5892c50a69ea45bfb","after":"f436aa1e7adfbe99422d03f9a398cd6b0de8c56a","ref":"refs/heads/main","pushedAt":"2024-04-23T01:00:46.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Text Analysis \n\nUsing NLP - bag of words approach and TF-IDF to predict news articles headlines as either positive (1) or negative (0)","shortMessageHtmlLink":"Text Analysis"}},{"before":"f21b5e0d5d9991eea83d92e589320bd7b827e808","after":"feef420eee099969bc83dbb5892c50a69ea45bfb","ref":"refs/heads/main","pushedAt":"2024-04-23T00:55:31.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Loan default prediction using SVM.ipynb\n\nPart A: \r\nThe goal of the first part is to build a predictive model based on Support Vector Machine to predict loan default probability. The data used for this project was gitten from LendingClub. \r\n \r\na. Use the main (unbalanced) sample to train a SVM classifier. Plot the confusion matrix and calculate the prediction evaluation metrics—i.e., accuracy, precision, and recall rates. \r\nb. Make a balanced sample by under-sampling the majority class and train a second SVM classifier on this new balanced sample. Show the model performance by calculating the confusion matrix and evaluation metrics. \r\nc. How did the change in the sample balance affect the prediction performance? \r\n\r\nPart B: Use decision trees, Random Forest, Gradient boosting to predict high earner given the rest of the details\r\nOne of the common problems with individual predictive models is that they are prone to overfitting. We want to build ensemble models to decrease the change of overfitting and predict an individual’s income class. These data are extracted from the 1994 Census database and consist of individual characteristics, such as age, education, marital status, occupation, among others. The last column of the dataset is a discrete variable, Salary, and contains one of two strings to indicate if the salary was below or above $50,000. Use this variable to make a categorical outcome variable, “HighIncome” that takes values of 0 and 1. \r\n\r\n1.\tBuild a Decision Tree classification model for individual’s income level. Use model predictions on the test sample to create a confusion matrix and calculate the accuracy score, precision, and recall ratios. \r\n2.\tBuild a Random Forest predictive model for individual’s income level. Use model predictions on the test sample to create a confusion matrix and calculate the accuracy score, precision, and recall ratios. \r\nc. Build a Gradient Boosting predictive model for individual’s income level. Use model predictions on the test sample to create a confusion matrix and calculate the accuracy score, precision, and recall ratios. \r\nd. Compare the performance of the Decision Tree model with each of the ensemble models. Can you explain the differences based on each model’s objectives?","shortMessageHtmlLink":"Loan default prediction using SVM.ipynb"}},{"before":"a22349dc4e79763b3f3f2efa177e896b2f5b9584","after":"f21b5e0d5d9991eea83d92e589320bd7b827e808","ref":"refs/heads/main","pushedAt":"2024-04-23T00:43:12.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Predict if an individual is a high earner\n\nThis project aims to build models (Logit, KNN, SVM) that predict whether an individual is a high-earner by observing some of their characteristics. The data is extracted from the 1994 Census database and consist of individual characteristics, such as age, education, marital status, occupation, race, sex, and salary, among others. \r\nThe last column of the dataset is a discrete variable, Salary, and contains one of two strings to indicate if the salary was below or above $50,000. Use this variable to make a categorical outcome variable, “HighIncome” that takes values of 0 and 1. \r\n\r\nUse 70% of the data to train your models and the rest as your test sample. Also, use age, Education level (numeric), hours per week, capital gain, and sex as your input features. [Hint: for some of the models you may need to manually encode other categorical variables such as sex as well.] \r\na) Build a Logistic Regression predictive model for individual’s income level. Use model predictions on the test sample to create a confusion matrix and calculate the accuracy score, precision, and recall ratios. \r\nb) Build a K-Nearest Neighbors classification model for individual’s income level. Use model predictions on the test sample to create a confusion matrix and calculate the accuracy score, precision, and recall ratios. \r\nc) Build a Support Vector Machine classifier for individual’s income level. Use model predictions on the test sample to create a confusion matrix and calculate the accuracy score, precision, and recall ratios. \r\nd) Rank the models based on their performance. Can you explain the differences in model performances based on each model’s pros and cons?","shortMessageHtmlLink":"Predict if an individual is a high earner"}},{"before":"a5875f760bcc6ca4db7bef64574c077dbcbca4de","after":"a22349dc4e79763b3f3f2efa177e896b2f5b9584","ref":"refs/heads/main","pushedAt":"2024-04-23T00:33:21.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Clustering and Dimensionality Reduction\n\n\r\nAsset management and investment allocation is a tedious and time-consuming process, in which investment managers often must design customized approaches for each client.  What if we were able to organize these clients into particular investor profiles, or clusters, where each group includes investors with similar characteristics? This helps with simplicity and standardization in the investment management process.\r\n\r\nWe want to build a clustering model to group investors based on parameters related to the \r\nability and willingness to take risk. \r\n\r\nThe data used summarizes survey responses for 3866 individual investors and includes 12 features for each person (Disclaimer: the original survey data has 10,000+ observations and 500+ features; I used a small part of the data for the purpose of this assignment). Note that this is an excel file; import with pd.read_excel().\r\n\r\nThe 12 features in the file include \r\nAGE (age)\r\nEDUC (education) \r\nMARRIED (marital status)\r\nKIDS (number of children)\r\nLIFECL (lifecycle stage)\r\nOCCAT (occupation)\r\nRISK (willingness to take risk)\r\nHHOUSES (house ownership)\r\nWSAVED (spending versus income)\r\nSPENDMOR (willingness to spend)\r\nNWCAT (net worth), INCCL (income)","shortMessageHtmlLink":"Clustering and Dimensionality Reduction"}},{"before":"bff6180f99fd9d9a968640c8de4bf29cc506e9b3","after":"a5875f760bcc6ca4db7bef64574c077dbcbca4de","ref":"refs/heads/main","pushedAt":"2024-04-23T00:27:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Vector AutoRegression (VAR)\n\nThe goal of this project is to evaluate the effect of monetary policy on inflation and unemployment rates in the US","shortMessageHtmlLink":"Vector AutoRegression (VAR)"}},{"before":"4f8dddf7b1863e39ebb0f582a33c873bd76ecb21","after":"bff6180f99fd9d9a968640c8de4bf29cc506e9b3","ref":"refs/heads/main","pushedAt":"2024-04-23T00:16:44.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Auto Regression and Partial Autocorrelation\n\nThis project estimates the Auto Regression and Partial Autocorrelation, AR(p), MA(q) and ARMA(p,q) for the S&P500 index returns","shortMessageHtmlLink":"Auto Regression and Partial Autocorrelation"}},{"before":"55ed6462a4b706a73fe2298882b78909219db7e3","after":"4f8dddf7b1863e39ebb0f582a33c873bd76ecb21","ref":"refs/heads/main","pushedAt":"2024-04-22T23:09:54.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Global Minimum variance portfolio analysis\n\nThis project focuses on three things: \r\n\r\na) Picking ten industries and plot the efficient frontier using those industries as the universe of risky assets.\r\nb) Find the asset weights, expected return, and volatility for the Global Minimum Variance Portfolio.\r\nc) Find the asset weights, expected return, and volatility for the Maximum Sharpe Ratio portfolio portfolio.","shortMessageHtmlLink":"Global Minimum variance portfolio analysis"}},{"before":"bc88ba8e6850be230998097708f0416f392d8b11","after":"55ed6462a4b706a73fe2298882b78909219db7e3","ref":"refs/heads/main","pushedAt":"2024-04-22T22:47:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"VaR and Efficient Frontier Analysis","shortMessageHtmlLink":"VaR and Efficient Frontier Analysis"}},{"before":"f51ffeb37bc332305b9bfa8a45a8f46755258430","after":"bc88ba8e6850be230998097708f0416f392d8b11","ref":"refs/heads/main","pushedAt":"2024-04-22T22:28:25.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Delete Homework 10 - Clustering & Dimension reductions.ipynb\n\ndelete file","shortMessageHtmlLink":"Delete Homework 10 - Clustering & Dimension reductions.ipynb"}},{"before":"232b685fb972d833b8ab7dcbfbb0c0c2bbc4dddd","after":"f51ffeb37bc332305b9bfa8a45a8f46755258430","ref":"refs/heads/main","pushedAt":"2024-04-22T22:27:47.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"This project analyzes returns from stocks","shortMessageHtmlLink":"This project analyzes returns from stocks"}},{"before":"21ff73aa8082bb79091fabb1de8016430ca87731","after":"232b685fb972d833b8ab7dcbfbb0c0c2bbc4dddd","ref":"refs/heads/main","pushedAt":"2024-04-22T19:53:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Clustering using K-means and DBSCAN","shortMessageHtmlLink":"Clustering using K-means and DBSCAN"}},{"before":null,"after":"21ff73aa8082bb79091fabb1de8016430ca87731","ref":"refs/heads/main","pushedAt":"2024-04-22T19:52:04.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Agoons20","name":"Duchoma Agoons","path":"/Agoons20","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/56331600?s=80&v=4"},"commit":{"message":"Initial commit","shortMessageHtmlLink":"Initial commit"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"Y3Vyc29yOnYyOpK7MjAyNC0wNS0wOVQwNjo1NDowNi4wMDAwMDBazwAAAARFVZrD","startCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wNS0wOVQwNjo1NDowNi4wMDAwMDBazwAAAARFVZrD","endCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wNC0yMlQxOTo1MjowNC4wMDAwMDBazwAAAAQ3Yi3-"}},"title":"Activity · Agoons20/Portfolio-Mgmt-And-Machine-Learning-in-Finance"}