https://github.com/loccx78vn/modeltime_in_r

This is a tutorial for R-user learn about package modeltime and how to use it to apply for time series forecast
https://github.com/loccx78vn/modeltime_in_r

Last synced: 21 days ago
JSON representation

This is a tutorial for R-user learn about package modeltime and how to use it to apply for time series forecast

• Host: GitHub
• URL: https://github.com/loccx78vn/modeltime_in_r
• Owner: Loccx78vn
• Created: 2024-11-19T07:05:43.000Z (about 2 months ago)
• Default Branch: main
• Last Pushed: 2024-11-19T09:39:01.000Z (about 2 months ago)
• Last Synced: 2024-11-19T10:40:11.799Z (about 2 months ago)
• Language: R
• Homepage: https://loccx78vn.github.io/modeltime_in_R/
• Size: 21.7 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Forecasting_time_series_2:

In the chapter 1 - "**Forecasting_time_series**", I introduced about the definition of time series, about a lot of forecasting approaches to deal with planning tasks. Our aim is analyzing time series by using machine learning.

Certainly! In R, both `timetk` and `modeltime` are packages designed to facilitate time series analysis and forecasting. Here's a brief summary of each:

# In R:

In this chapter, I will present the tutorial of time series forecasting by package `modeltime` and the combination between packages `timetk` and `recipe`.

## `timetk`

- **Purpose**: `timetk` is used for time series data manipulation and visualization. It helps in handling and transforming time-based data to make it more accessible for analysis and forecasting.

- **Key Features**:

  - **Date/Time Parsing**: Easily convert and format dates and times.

  - **Feature Engineering**: Create time-based features (e.g., year, month, day, weekday) to use in models.

  - **Visualization**: Functions for plotting time series data, including plotting with `ggplot2` and creating interactive visualizations.

  - **Aggregation and Resampling**: Aggregate data at different time intervals and resample time series data for different frequencies.

## `modeltime`

- **Purpose**: `modeltime` is designed for forecasting with time series data. It integrates with the `tidymodels` framework, enabling users to build, tune, and evaluate time series forecasting models.

- **Key Features**:

  - **Model Building**: Provides functions to fit various forecasting models, including ARIMA, ETS, and machine learning models.

  - **Model Evaluation**: Tools for assessing model performance using cross-validation and other metrics.

  - **Model Ensembling**: Combine multiple models to improve forecasting accuracy.

  - **Integration**: Works seamlessly with `timetk` for time series data preparation and `parsnip` for model specification.

In summary, `timetk` is great for preparing and visualizing time series data, while `modeltime` is focused on the modeling and forecasting aspects, allowing you to build and evaluate time series models effectively.

# Conclusion:

I hope you can enjoy the pratice and hope you have a great day!!! See you soon in another future topic.