UCLA

In this analysis, the objective is to forecast the stock prices of property and casualty insurance in 2022. This industry is known to be relatively stable and resilient to economic downturns. The data utilizes weekly adjusted closing prices of Progressive Insurance from 2019 to 2021 to form the training set. Three different models were created to predict weekly adjusted closing prices for 2022. The methods used were the LSTM and GRU recurrent neural network models, as well as the ARIMA time series analysis. Based on the results, the GRU method achieved the lowest RMSE due to its ability to avoid overfitting and does not rely on the assumption of stationarity.