I am student conducting an experiment with different models for time series prediction. In my experiment, I am going to use ARIMA, a Recurrent Neural Network, a Long-Short Term Memory network, and a Temporal Convolutional Neural network.

The dataset is from a store and the dependent variable is number of visitors. As independent variables I have things like average daily temperature (and many others). Also, there is a time dimension with the scale one day per unit.

I also want to build a markov model for this project (for learning purposes and they seem interesting). I have been studying Markov chains and Markov Chain Monte Carlo simulations. However, neither really seem to be suitable for my dataset based on my limited understanding.

What type of markov model should I use/investigate further if my dataset has number of visitors as dependent variable, a time dimension and several independent variables not related to time?

Is it feasible to use a markov model on this kind of problem?

  • 1
    $\begingroup$ Do you mean something more specific by a Markov model? A Markov model typically refers to a model satisfying the Markov property that the future state is independent of the past state given the present state. If you define the state variables properly, an ARIMA model (and many others) satisfy the Markov property. $\endgroup$ – Matthew Gunn Feb 22 at 20:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.