Description

On this course, you’ll grow to be conversant in a wide range of up-to-date monetary evaluation content material, in addition to algorithms methods of machine studying in the Python surroundings, the place you’ll be able to carry out extremely specialised monetary evaluation. You’ll get acquainted with technical and elementary evaluation and you’ll use completely different instruments on your evaluation. You’ll get acquainted with technical and elementary evaluation and you’ll use completely different instruments on your evaluation. You’ll study the Python surroundings utterly. Additionally, you will study deep studying algorithms and synthetic neural networks that may tremendously improve your monetary evaluation expertise and experience.

This tutorial begins by exploring numerous methods of downloading monetary information and making ready it for modeling. We test the essential statistical properties of asset costs and returns, and examine the existence of so-called stylized details. We then calculate well-liked indicators used in technical evaluation (reminiscent of Bollinger Bands, Shifting Common Convergence Divergence (MACD), and Relative Energy Index (RSI)) and backtest computerized buying and selling methods constructed on their foundation.

The subsequent part introduces time collection evaluation and explores well-liked fashions reminiscent of exponential smoothing, AutoRegressive Built-in Shifting Common (ARIMA), and Generalized Autoregressive Conditional Heteroskedasticity (GARCH) (together with multivariate specs). We additionally introduce you to issue fashions, together with the well-known Capital Asset Pricing Mannequin (CAPM) and the Fama-French three-factor mannequin. We finish this part by demonstrating alternative ways to optimize asset allocation, and we use Monte Carlo simulations for duties reminiscent of calculating the value of American choices or estimating the Worth at Danger (VaR).

Within the final a part of the course, we supply out a whole information science venture in the monetary area. We strategy bank card fraud/default issues utilizing superior classifiers reminiscent of random forest, XGBoost, LightGBM, stacked fashions, and many extra. We additionally tune the hyperparameters of the fashions (together with Bayesian optimization) and deal with class imbalance. We conclude the guide by demonstrating how deep studying (utilizing PyTorch) can clear up quite a few monetary issues.

If the coupon is just not opening, disable Adblock, or attempt one other browser.

Leave a comment

Your email address will not be published. Required fields are marked *