momentum strategy based on trend

Hi, I have a dataset as attached. The columns represent stocks, the rows representing time. The values are the lag 3 moving average prices of the 10 stocks. As you can see, some data are missing at the beginning or in the end of the time horizon, this is because the stocks are either not listed or have been delisted and returns are not available. The basic idea is to regress cross-sectionally monthly prices at date t on all moving average series at date t-1, finally predict monthly prices at date t+1 using the regression estimates and the moving average series at date t. This procedure guarantees we forecast stock prices at t+1 with information set only up to t. We then rank all stocks based on the forecasts into five quintiles( 2 stocks in each quantile,treated as a equally weighted portfolio), long the quintile with highest forecast prices and short the quintile with lowest, and rebalance in three month.Therefore, the output should be the 3-month return vector of the long portfolio and a vector of the short portfolio. ... I tried with the ARMA to forecast but then I didn't manage to sort and get the long and short portfolios. Thank you for your help.