Dear all,
I find myself in a bit of a predicament and could use some guidance. I'm currently working on building a model to forecast the excess returns of a REIT ETF. Initially, I considered using an ARMA model. However, after conducting tests for stationarity, I've found that none of the partial autocorrelation (PAC) or autocorrelation (AC) values are significant, indicating that no lags are particularly "useful" to include in the model.
This has left me questioning whether the ARMA model is the right choice for this task. I would greatly appreciate any advice or suggestions regarding alternative models that could be more suitable for forecasting stock returns. I've considered GARCH ( but that allows only to model the volatility).
Thank you in advance for any assistance you can provide.
I find myself in a bit of a predicament and could use some guidance. I'm currently working on building a model to forecast the excess returns of a REIT ETF. Initially, I considered using an ARMA model. However, after conducting tests for stationarity, I've found that none of the partial autocorrelation (PAC) or autocorrelation (AC) values are significant, indicating that no lags are particularly "useful" to include in the model.
This has left me questioning whether the ARMA model is the right choice for this task. I would greatly appreciate any advice or suggestions regarding alternative models that could be more suitable for forecasting stock returns. I've considered GARCH ( but that allows only to model the volatility).
Thank you in advance for any assistance you can provide.
