Define a dependent variable that takes the values "up" or "down", representing the direction of the price move of the S\&P 500 Index for the following day from the previous day. Now define regressors that represent the difference in the current (closing) price and the 50-day moving average and the difference in the current (closing) price and the 200-day moving average.

Obtain the daily closing of the S\&P 500 for the year 2018, and as much of 2017 as necessary to compute 200-day moving averages. Use the 2018 data to fit a logistic regression model with the up or down closes of the following day as the response variable. (Notice that response variable is for the day following the corresponding regressor variables.)

Test the fit of your model for the first 10 trading days of 2019 .