The data given is of the mutual funds in the USA.
The objective of this problem is to predict the ‘basis point spread’ over AAA bonds i.e. feature ‘bonds_aaa’ against each Serial Number.
Basis Point Spread indicates the additional return a mutual fund would give over the AAA-rated bonds.
Fund Symbol - Uniques symbol for the mutual fund used for representing it on the bourses
Fund Name - Full name of the mutual fund scheme
Category - Investment category of the mutual fund
Fund Family - Asset management company to which mutual fund belongs to
Investment - Type of investment of the mutual fund scheme
Size - Size of the mutual fund based on the total net assets
Total net assets - Total assets under management for the mutual fund scheme
Currency - Currency in which the investments of the mutual fund are held
Net Annual Expense Ratio - Expense ratio is the fee that the asset management company charges to the clients as a percentage of the total assets.
Morningstar Rating - This is the overall fund rating given by the rating agency Morning Star. The rating is on the scale of 1 to 5 where 5 is the best.
Inception Date - The date on which the mutual fund scheme was started.
portfolio - percentage of total assets invested in the investment instrument.
sectors - percentage of equity assets invested in the sector
Morningstar Return Rating - Fund rating based on returns by the rating agency Morning Star.The rating is on the scale of 1 to 5 where 5 is the best. Returns_ytd: Year to date return of the mutual fund.
retruns - Annual return of the mutual fund for the respective year
Morningstar Risk Rating - Fund rating based on risk of the mutual fund by the rating agency Morning Star.The rating is on the scale of 1 to 5 where 5 is the best.
Alpha 3y - 3year average alpha of the mutual fund.
Beta 3y - 3year average beta of the mutual fund.
Mean Annual Return 3y - 3year mean annual return
Standard Deviation 3y - Standard deviation of returns over three years.
Sharpe Ratio 3y - 3year average Sharpe ratio of the mutual fund.
bonds_* (Target) - Basis point spread over the bonds for the mutual fund.
For this particular dataset we are using RMSE as the evaluation metric.
After completing this project I have better understanding of how to apply linear model using GridsearchCV.
Linear regression
GridsearchCV
Ridge and Lasso Regressor