| Title: | Execution of Time Series Models |
|---|---|
| Description: | Execution of various time series models and choosing the best one either by a specific error metric or by picking the best one by majority vote. The models are based on the "forecast" package, written by Prof. Rob Hyndman. |
| Authors: | Avi Blinder <[email protected]> |
| Maintainer: | Avi Blinder <[email protected]> |
| License: | GPL-2 |
| Version: | 0.1.8 |
| Built: | 2024-11-10 05:15:06 UTC |
| Source: | https://github.com/aviblinder/tsselect |
Internal function that verifies the class of the object (should be time series)
check_object(x)check_object(x)
x |
A timeseries object |
internal function for verifying that the object belongs to class "time series"
stops if object not a ts class
Avi Blinder
This sample dataset is taken from a Kaggle's competition ("Rossman Store Sales")
data("ros1_ts")data("ros1_ts")
The format is: Time-Series [1:365] from 1 to 365: 0.1 5530 4327 4486 4997 ...
Only 2013 sales data from the first store are represented in the dataset
"https://www.kaggle.com/c/rossmann-store-sales/data"
data(ros1_ts) ## maybe str(ros1_ts) ; plot(ros1_ts) ...data(ros1_ts) ## maybe str(ros1_ts) ; plot(ros1_ts) ...
This sample dataset is taken from a Kaggle's competition ("Rossman Store Sales")
data("ros2_ts")data("ros2_ts")
The format is: Time-Series [1:365] from 1 to 365: 0.1 5530 4327 4486 4997 ...
Sales data from the second store are represented in the dataset
"https://www.kaggle.com/c/rossmann-store-sales/data"
data(ros2_ts)data(ros2_ts)
Function that executes several models and picks the best one.
run_models(ts1, accuracy_measure = NULL)run_models(ts1, accuracy_measure = NULL)
ts1 |
A timeseries object |
accuracy_measure |
- Possilbe error meassures: ME, RMSE, MAE, MPE ,MAPE, MASE, ACF1 |
the best time series model
Avi Blinder
data(ros1_ts) run_models(ros1_ts) run_models(ros1_ts,"RMSE")data(ros1_ts) run_models(ros1_ts) run_models(ros1_ts,"RMSE")