Using data from the excel file and coding in Python, you should now estimate the following: for each ETF, estimate the sensitivity of ETF flows to past returns. a. Write down the main regression specification, and estimate at least five regression models based on it (e.g., with varying the number of lags). Then, present the regression output for one ETF of choice, including coefficients with t-stats, R squared, and number of observations.<\/p>\n
a. Estimate the OLS regression from (2a) for each ETF and save betas. Then, conduct cluster analysis using k-means clustering with different variables, but for a start, try these two dimensions: i. Flow-performance sensitivity (i.e., betas from point (2)) vs fund size (AUM). ii. Propose at least one other dimension, and perform the cluster analysis again. What did you learn? iii. Now, instead of clustering, analyse fund types, and see whether flow- performance sensitivity varies by fund type.<\/p>\n
dm me so that I can send you the cleaned up data<\/p>\n<\/div>\n
submitted by \/u\/Spiritual_Key_2204 <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
