Choosing the right Time Series Forecasting Methods

Choosing the right Time Series Forecasting Methods

There are different time series forecasting methods to forecast stock price, demand etc. In this machine learning project, you will learn to determine which forecasting method to be used when and how to apply with time series forecasting example.

Videos

Each project comes with 2-5 hours of micro-videos explaining the solution.

Code & Dataset

Get access to 50+ solved projects with iPython notebooks and datasets.

Project Experience

Add project experience to your Linkedin/Github profiles.

What will you learn

What is a Time Series?
How to create a time series?
Visualizing a time series plot
How to convert a multiplicative time series into an additive time series
Logarithmic transformation its significance and its use
Assumptions of a time series forecasting model
To extract the trend, seasonality and random terms from the model
How to decompose a time series
Using standard regression-based methods to do forecast
How to identify autocorrelation and partial autocorrelation
How to de-trend a time series
Augmented Dickey-Fuller Test (ADF test for stationarity)
When and how to use the differencing of a time series
Double exponential smoothing model
MAPE metric to compare the accuracy of any time series model
What is the ARIMA model and steps to initialize model
Applying ARIMA model for forcasting

Project Description

In this machine learning project, we will be taking open source datasets that are publicly available and will be discussing various methods/techniques of performing time series forecasting. We will discuss about the traditional methods such as holt-winters method, Autoregressive integrated moving average method, exponential smoothing methods, as well we will also be comparing the modern methods of performing forecasting using neural network based models.

 

Similar Projects

In this data science project, you will predict borrowers chance of defaulting on credit loans by building a credit score prediction model.

Build a predictive model to correctly classify products between 9 product categories (fashion, electronics, etc.) using the Otto Group dataset.

In this project, we are going to work on Deep Learning using H2O to predict Census income.

Curriculum For This Mini Project

What is Time Series
04m
How to create Time Series in R
05m
Components of Time Series
03m
Assumptions of Time Series Forecasting
05m
Investigate components of Time Series
02m
Example - Stock Market Dataset
01m
Decompose method - Additive/Multiplicative Component
03m
STL method
02m
Plot graph for Decompose and STL function
01m
How to do Time Series Forecasting
08m
Create Time Series Lag
10m
Add more Lag variables
02m
Identify AutoCorrelation
11m
Partial Autocorrelation
01m
DeTrend a Time Series using Johnson method
07m
Deseasonalize Time Series
05m
Seasonality Plot
03m
Augmented Dickey Fuller (ADF) Test
03m
KPSS Test
02m
Air Passenger Data - ADF Test
01m
Differencing Method
17m
Moving Average Method
04m
Data Visualization
01m
Forecast Methods
03m
Moving Average Method
06m
Exponential Smoothing Method
01m
Simple Exponential Smoothing Method
06m
Double Exponential Smoothing Method
01m
Triple Exponential Smoothing Method
01m
Compare accuracy of models
07m
Generate forecast using all 3 methods
03m
Arima - Auto Regressive Integrated Moving Average Model
17m
Get the Dataset ready
07m
Decompose Time Series Data
08m
Diffrentiation
05m
How to identify Lags
05m
Auto Arima
15m
Forecast
02m
Dummy Variables & Box Cox Transformation
02m
Neural Network based Forecasting
09m
Hyperparameter Tuning to increase accuracy
06m
Conclusion
01m