In forecasting the performance of many agents in a time series, there is a strong seasonality component, in addition to non-seasonal features for each agent. How can I capture the overall seasonal trends in my model building process without overfitting? Currently, I’m using a time-series cross validation approach to fitting a Random Forest Regressor or … Read more
I am learning on how to predict with ARIMA models. To get some knowledge I read trough some online tutorials for R and ARIMA models. Now I wanted to try this by myself with a problem I am currently working on. The goal is to predict the vehicle speed of a car based on past … Read more
If I have an estimate for Holt Winters model as the attached image. How do I interpret the estimates i.e the level, trend and seasonal smoothing weight. Answer I think these are the alpha, beta and gamma coefficient for the model. Basically the model is to be written as: If you are asking the “meaning” … Read more
Wouldn’t a better scaling factor be with the MAE produced by a naive forecast on the test data itself? When evaluating MASE for the training set, this essentially becomes a comparison for the forecast model with a naive one, why do we not take this approach with the test set? Answer AttributionSource : Link , … Read more
Here is what my notes say about estimation and prediction: Estimating the conditional mean We need to estimate the conditional mean $\beta_0+\beta_1x_0$ at a value $x_0$, so we use $\hat{Y_0}=\hat{\beta_0}+\hat{\beta_1}x_0$ as a natural estimator. Here we get $$ \hat{Y_0} \sim N\left(\beta_0+\beta_1x_0,\sigma^2h_{00}\right) \,\,\,\,\,\,\,\,\,\,\,\ \text{where} \,\,\,\,\,\,\,\,\,\,\,\ h_{00} = \frac{1}{n}+\frac{(x_0-\bar{x})^2}{(n-1)s_x^2} $$ with a confidence interval for $E(Y_0) =\beta_0+\beta_1x_0$ … Read more
I have the code below which trains a model with some predictors, forecasts it one step, appends the forecasted value on the original training data and then tries to feed that back in and train and forecast another model ahead one step. The first time it trains and forecasts the model works fine, but the … Read more
If i’m working in a car company and I have some data for every customer i.e. Their license plate Date of their car’s service in the dealer Number of km in their car when they service it Their periodic service (e.g. 1000 km, 2000 km, etc.) Can we predict when will the customer back to … Read more
I am exploring the use of ARIMA and Seasonal ARIMA models (SARIMA). In some of my datasets, I can clearly observe seasonality in the ACF and PACF plots (the lines at seasonal lags clearly cutting the confidence interval region). In order to make such data stationary, and to account for seasonality in my model, is … Read more
I have been working with a finance team to help forecast revenue for some product data. Particularly when the series are short and difficult to forecast, their first response is to add a bunch of “driver” data (basically external regressors) to the model. I have given presentations a few times on overfitting and why adding … Read more
I have a data for each minute as shown below Timestamp cpu-usage 2017-02-14 00:00:00 1.80000 2017-02-14 00:01:00 16.04000 2017-02-14 00:02:00 23.16000 2017-02-14 00:03:00 24.21400 2017-02-14 00:04:00 24.74100 2017-02-14 00:05:00 12.88767 I have created a graph for this data >plot(data.ts) I have even plot Correlation plot for this >acf((log(data.ts))) >pacf((log(data.ts))) Now by seeing the above graphs, … Read more