Scribd Logo
Upload

Welcome to Scribd!

  • 27 views

    CQF June 2021 M4L4 Exercises

    Uploaded by

    PrathameshSagare
    This document outlines two machine learning exercises. The first asks to perform feature selection on the Boston housing dataset using various techniques and compare regressor performance. The second asks to create a custom transformer to replace outlier values in the S&P 500 percentage return dataset and compare results to the original feature set. Both exercises are meant to demonstrate feature selection and outlier treatment skills using scikit-learn and real-world datasets.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    CQF June 2021 M4L4 Exercises

    Uploaded by

    PrathameshSagare
    27 views1 page
    This document outlines two machine learning exercises. The first asks to perform feature selection on the Boston housing dataset using various techniques and compare regressor performance. The second asks to create a custom transformer to replace outlier values in the S&P 500 percentage return dataset and compare results to the original feature set. Both exercises are meant to demonstrate feature selection and outlier treatment skills using scikit-learn and real-world datasets.

    Original Title

    CQF_June_2021_M4L4_Exercises

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document outlines two machine learning exercises. The first asks to perform feature selection on the Boston housing dataset using various techniques and compare regressor performance. The second asks to create a custom transformer to replace outlier values in the S&P 500 percentage return dataset and compare results to the original feature set. Both exercises are meant to demonstrate feature selection and outlier treatment skills using scikit-learn and real-world datasets.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    27 views1 page

    CQF June 2021 M4L4 Exercises

    Uploaded by

    PrathameshSagare
    This document outlines two machine learning exercises. The first asks to perform feature selection on the Boston housing dataset using various techniques and compare regressor performance. The second asks to create a custom transformer to replace outlier values in the S&P 500 percentage return dataset and compare results to the original feature set. Both exercises are meant to demonstrate feature selection and outlier treatment skills using scikit-learn and real-world datasets.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 1

    Machine Learning Exercises - I

    September 13, 2021

    Kannan Singaravelu
    [email protected]
    ∗∗∗

    Exercise 1
    For the Boston housing dataset, construct a dataframe with all features and target values. Perform
    feature selection to choose the most appropriate features using
    1. Variance Inflation Factor
    2. SelectKBest
    3. Recursive Feature Elimination
    4. Recursive Feature Elimination Cross validation
    5. Shapley Additive Explanations
    Fit the regressor and compare the results. How much better does the model perform? Use Scikit-
    learn package to perform this task.

    Exercise 2
    Create a custom transformer that replaces outlier values of 1, 5, 20, 60 and 120 days SPX percentage
    returns. Determine the lower and upper bound of acceptable values based on the q-th percentile.
    Compare the result with the original feature set. Use the SPX dataset used in the Python Labs.
    —————————
    Kannan Singaravelu
    Certificate in Quantitative Finance, June 2021

    You might also like