ARIMA (Autoregressive Integrated Moving Average) Analysis

Financial and operational decisions are made based on current market conditions and predictions of how the future looks. Past data is aggregated and analyzed to find patterns, & used to predict future trends and changes. Forecasting allows your company to be proactive instead of reactive.

Forecasting the Data

The Datasets were made randomly and named them Seven_Plus mobile and M&H Clothing.

It is important to Analyse data before forecasting as it helps in understanding the current position of the business/organisation. It is also crucial to understand the post situations of the Organisation. So, I have done an Exploratory Data Analysis using pyhton3

Click here to see the Exploratory Data Analysis.

But, have you ever thought — How a Forecasting system really works?

I am sure you have thought of it, so let me help you with it.

This system uses the ARIMA technique. I have prepared a model using the Orange — Data Mining Tool.

Orange?

Orange is a visual programming software package used for this domain. It has been used widely, ranging from machine learning, data mining, and data analysis, etc. Orange tools (called widgets) are within the realm of simple data visualization & pre-processing empirical evaluation of learning algorithms and predictive modelling. Visual programming is implemented via a combination in which workflows are designed by linking user-designed widgets.

At the same time, proficient users can use Orange as a Python library to manipulate data and alter widgets.

AFTER DATA ANALYSIS, ARIMA MODEL WAS USED TO DO THE FORECASTING.

Let me show you how ARIMA works:

.CSV file for Clothing Dataset

.CSV file for SEVEN PLUS Dataset

PERFORMING ARIMA MODEL ON SEVEN PLUS DATASETS

Firstly, the .csv file was uploaded in the model using Data Widget, and Quality Rating was taken as the Target Value.

Then ‘As Timeseries’ widget was selected to make the data stationary to make the ARIMA Model.

This widget reinterprets any data table as a time series to be used with the rest of the widgets in this add-on. In the widget, you can set which data attribute represents the time variable.

The third step was putting the Data into ARIMA Model (Autoregressive Integrated Moving Average)

Using this widget, you can model the time series with the ARIMA model.

1. Model’s name. By default, the name is derived from the model and its parameters.

2. Use exogenous data. Using this option, you need to connect additional series on the Exogenous data input signal.

3. Number of forecast steps the model should output, along with the desired confidence intervals values at each step. The confidence level was set at 95%.

The Final step was selecting the Data Table Widget.

The Data Table widget receives one or more datasets in its input and presents them as a spreadsheet. Attribute values may sort data instances. The widget also supports the manual selection of data instances.

Similarly, this model was performed on all the other two important aspects of the Seven Plus Dataset, i.e., Durability and Usage.

PERFORMING ARIMA MODEL ON CLOTHING DATASET

Firstly, the .csv file was uploaded in the model using Data Widget, and Quality Rating was taken as the Target Value.

Then again, ‘As Timeseries’ Widget and ‘ARIMA’ Widget was used as shown above.

In the Seven Plus Dataset, the Forecasting Data can be observed as:

1- On the Basis of Quality rating

2- On the Basis of Usage (on a Daily Basis)

3- On the Basis of Durability (Months)

So, it can be said that on an average:

1. the next Quality rating would be around — (6.56; 7.38 and 6.91)

2. the next Usage (On a daily basis) would be around — (14.98; 12.52 and 4.18)

3. the next Durability (monthly) would be around — (53.05; 54.12 and 54.18)

Graph of Forecasting can be viewed as:

P.S. — Unknown people are the Future Customers on whom forecasting was done.

From this graph, it can be observed that — In future,

1. Durability becomes more stable.

2. Usage will increase at a diminishing rate.

3. Quality will also become more stable

In the Clothing Dataset, the Forecasting Data can be observed as:

1- On the Basis of Quality rating

2- On the Basis of Durability (Monthly)

So, it can be said that on an average:

1. the next Quality rating would be around — (5.37; 5.86 and 5.72)

2. the next Durability (Monthly) would be around — (3.13; 3.44 and 3.49)

Graph of Forecasting can be viewed as:

P.S. — Unknown people are the Future Customers on whom forecasting was done.

From this graph, it can be observed that — In future,

1. Durability become more stable.

2. Quality will also become more stable

In this way, ARIMA works.

Contacts

In case you have any questions or any suggestions on what my next article should be about, please leave a comment below or mail me at aryanbajaj104@gmail.com.

If you want to keep updated with my latest articles and projects, keep visiting the website ^_^.

Connect with me via:

LinkedIn