gmwm
gmwm copied to clipboard
Generalized Method of Wavelet Moments (GMWM) is an estimation technique for the parameters of time series models. It uses the wavelet variance in a moment matching approach that makes it particularly...
**Issue:** We need to support the introduction of the new TS object (c.f. #36) within graphs and object generators **Solution:** 1. Frequency should change the scale of the graph. If...
**Issue:** Per the new ability in #140 , the graph titles current only show "Haar Wavelet ... " **Solution:** Update the graph objects to use the filter attribute to construct...
**Issue:** Mapping is done primarily using `log()`, `logit()`, or a customized range version. In this case, we take a f:R->R^+ **Idea:** Perhaps a better alternative is to use the [hyperbolic...
**Feature** We currently only have support for the allan variance and the wavelet variance. It would be nice to add in the Hadamard and Total Variance. See: [Handbook of Frequency...
To do list: 1. Begin process of abstracting parts of the imu graphing procedure to align with multivariate time series. 2. Move out parameter checks into their own callable function....
Parallelization To speed up the bootstraps, we can parallelize different portions of the code. In order to do that, we need to enable stored seeds. This is a more concrete...
**Feature** `lts` acts as a great way to visualize what is happening within a latent time series. However, it only currently shows the breakdown of the process realizations. It would...
**Issue:** Right now, the GMWM is really restricted to people who understand R. We have to take into consideration the notion that engineers or other practitioners might not know R....
**Issue:** Currently, we are running suggested theta's through the GMWM estimator to obtain theta estimates. We have no way of checking initial starting values outside of this GMWM scope. **Solution:**...
**Feature:** Given a `ts.model` object filled in with parameters, it would be nice to export that into latex table form with appropriate spacing. E.g. AR1(phi=0.8, sigma2 = 0.01) + WN(sigma2...