RA- Deep Dive Into Forecasting – Excel and Python.

  • Post category:StudyBullet-14
  • Reading time:14 mins read


Forecasting with Excel & Python. Machine learning and statistical forecasting for Supply Chain.

What you will learn

Time Series Decomposition.

Univariate analysis for time series.

Bivariate analysis and auto-correlation.

Smoothing the time series.

seasonally adjusting the time series.

Generating and Calibrating Forecasting in Excel.

Learning Python and using it as everyday tool for forecasting.

Using the sktime Package for advanced forecasting methods and aggregations.

Time Series Forecasting.

Different Applications of forecasting.

Python

Arima

Machine learning forecasting

hierarchal forecasting

Excel

Description

Hello 🙂

Forecasting has been around for 1000s of years. it stems from our need to plan so we can have some direction for the future. We can consider forecasting as the stepping stone for planning. and that’s why it is as important as ever to have good forecasters in institutions, supply chains,  companies, and businesses.

With the ever-growing concerns of sustainability and Carbon-footprint. Would you believe it? a good forecast actually contributes to saving resources through the value chain and actually saving the planet. one forecaster at a time. needless to mention, forecasting is integral in marketing, operations, finance, and planning for supply chains…. pretty much everything

This course is aimed to orient you to the latest statistical forecasting techniques and trends. but first, we need to understand how forecasting works and the reasoning behind statistical methods, and when each method is suitable to be used.  that’s why we start first with excel and we scale with R. “Don’t worry if you don’t know Python, Crash fundamental sections are included!.

the course is for all levels because we start from Zero to Hero in Forecasting.

in this course we will learn and apply :

1- Time Series Decomposition in Excel and Python.

2- Univariate analysis for time series in Excel and Python..

3- Bivariate analysis and auto-correlation in Excel and Python..

4- Smoothing the time series and getting the Trend with Double and centered moving average.

5- seasonally adjusting the time series.

6- Simple and complex forecasts in Excel.

7- Use transformations to reduce the variance while forecasting.

8-Generating and Calibrating Forecasting in Excel.


Get Instant Notification of New Courses on our Telegram channel.


9- Learning Python  and using it as an everyday tool for forecasting.

10- Using the Fable Package for advanced forecasting methods and aggregations.

11- Using Forecast package for grid search on ARIMA.

12- Applying a workflow of different models in two lines of code.

13- Calibirating forecasting methods.

14- Applying Hierarchical time series with Bottom-up, middle out, and Top-down Approaches.

16-  Use the new R-Fable reconciliation method for aggregation.

15- Using Fable to generate forecasts for 10000  time-series and much more !!

*NOTE: Many of the concepts and analysis I explain first in excel as I find excel the best way to first explain a concept and then we scale up, improve and generalize with Python.. By the end of this course, you will have an exciting set of skills and a toolbox you can always rely on when tackling forecasting challenges.

Happy Forecasting!

Haytham

Rescale Analytics

Feedback from Clients and Training:

English
language

Content

Introduction

Introduction
Forecasting is the stepping stone of planning
Time Series
Difficulties in forecasting
Forecasting applications
Forecasting in inventory management
Different Forecasting Methods
2020 and COVID
Time Series analysis
Causal Methods
Stationarity of the data
Summary
Quiz on Chapter 1

Time Series and Pattern extraction

Introduction
Univariate Statistical analysis
Univariate Part2
Bivariate Statistics
Auto-Correlation
Assignment
Assignment Solution
Summary

Simple forecasting methods

Simple Forecasting methods
Naive and Seasonal Naive
Mean Percentage error
Seasonal average
Mean absolute scaled error
Simple exponential smoothing and log transformations
Simple forecasting Methods
Naive and Simple forecasting methods
linear Regression , Custom weighted moving average and SES
Optimizing the Parameters
Best Simple Forecasting Method
Simple Forecasting assignments
Solution
Summary

Double Moving average, Centered Moving average and Decomposition.

Introduction
Moving Averages
De-trending series
Time-series Decomposition
Additive Decomposition
Multiplicative Decomposition
Assignment
Decomposition Solved
Summary

Exponential Smoothing

Introduction
Simple Exponential Smoothing
Holt Exponential Smoothing
Initialization of alpha and Beta
Holt Model in Excel
Holt-winters Explanation
Additive Holt Winters Model
12 month Forecast with Holt Winters
Multiplicative Holt-Winters
12 Month ahead with multiplicative exponential smoothing
Assignment Holt
Assignment Solution

Multiple linear Regression

introduction
Intro to linear regression
Multiple linear regression in excel
Fitting the model
Shifting to Python

Welcome to Python

Python!
downloading Anaconda
Installing Anaconda
Spyder overview
Jupiter Notebook overview
Python Libraries
Summary

Python Programming fundmentals

Intro
Dataframes
Arithmetic Calculations with Python
Lists
Dictionaries
Arrays
Importing data in Python
Subsetting Data Frames
Conditions
Writing functions
mapping
for loops
for looping a function
Mapping On a data frame
for looping on a data frame
Summary
Assignment
Assignment answer 1
Assignment answer 2

working with dates in Python

Dates intro
datetime
Last purchase date and recency
recency histogram
Modeling inter-arrival time
Modeling inter-arrival time 2
Modeling inter-arrival time 3
Resampling
rolling time series
rolling Time series 2
Summary
Assignment
Assignment answer

Statistical Forecasting in Python

Introduction
Time Series Intro
Accuracy Measures
Preparing the data for time-series
Getting the time series components: Lecture
Getting the time series components
components uses
Arima Models
Stationarity test in python
Arima in python
ARIMA diagnostics
Grid search
For looping ARIMA
error handling
fitting the best model
Mean absolute error
Arima Comparison
Exponential smoothing
Exponential smoothing in python
Comparing exponential smoothing models
Time series summary
Assignment.
Assignment Explanation 1
assignment explanation 2
Assignment explanation 3
Assignment Explanation 4

Machine learning forecasting with sktime

Installing sktime
Why Forecasting is different from normal machine learning sklearn?
Different Fitting strategies with sktime
Different estimators in sktime
Libraries
Transforming from weekly to monthly timeseries
Changing from a normal date to a period date
Splitting timeseries
Knearestneighbor
Deriving the future
updating the time series with extra 2 years
Defining a forecast function
Transformed target Regressor
Testing the function
Plotting the results
Measuring acccuracy
Cross Validation
Conclusion
Assignment
Assignment Explanation part 1
assignment explanation part 2
Assignment explanation part 3
Assignment part 4
Assignment part 5
Assignment Part 6
Assignment last part
Summary

Hierarchal forecasting

Introduction
Levels of a Hierarchy
Middle-out approach
Top Down approach
Forecasting level Usage
Reconciliation
Tourism Data
Making Quarterly series
Indexing as a Hierarchy
Fitting Multiple models at once
Aggregations
Bottom up Forecasting
Top Down forecasting
Comparing Forecasts
Level 0 Comparison
Level 0 part 2
Topdown and weighted least squares
Final note
Enroll for Free

💠 Follow this Video to Get Free Courses on Every Needed Topics! 💠

Tags: ,