Skip to content
This project aims to understand trends in agricultural prices through quantity arrival data for different commodities in state of Maharashtra, India.
Jupyter Notebook
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.ipynb_checkpoints
Raw data data added Nov 5, 2018
Screenshots
Flagsetfluctuation.csv final Nov 14, 2018
.gitignore
CMO_MSP_Mandi_filtered.csv Visualizing and removing outliers Nov 6, 2018
Flagsetfluctuation.csv Identifying fluctuating commodity prices Nov 13, 2018
Monthly_data_cmo_filtered.csv
README.md
Task 2 (pre-processing).ipynb final Jan 21, 2019
Task 3 Compare prices.ipynb final Jan 21, 2019
Task- 2 Account Seasonality.ipynb final Jan 21, 2019
Task-1 Outlier detection.ipynb final Jan 21, 2019
Task-4. Flag set highest price fluctuation.ipynb
deseasonalize_data.csv
seasonal_data_analysis.csv
seasonality_type.csv final Nov 14, 2018

README.md

Time Series Analysis on agricultural commodity prices

This project aims to analyze trends and seasonality in datasets consisting of commodity prices for different APMCs of Maharashtra. Don't forget to star this repository if it helped you in any way. Thanks!

Project Status: [Completed]

Project Intro/Objective

The purpose of this project is identify relevant seasons of different agricultural commodities. It furthermore analyzes price fluctuations and flags the ones with most fluctuation in relevant months of the season.

Through this project, I've tackled the following tasks:

1. Testing and filtering outliers from the data.

2. Understanding price fluctuations accounting the seasonal effect

  • 1. Detecting seasonality type (multiplicative or additive) for each cluster of APMC and commodities
  • 2. De-seasonalising prices for each commodity and APMC according to the detected seasonality type

3. Comparing prices in APMC with MSP(Minimum Support Price)- raw and deseasonalised

4. Flagging set of APMC and commodities with highest price fluctuation across different commodities in each relevant season, and year.

Skills portrayed through this project

  • Time-series Analysis
  • Data Crunching
  • Inferential Applied Statistics
  • Outlier Detection through Inter-quartile ranges
  • Statistical Decomposition of Time-series data
  • Data Visualization

Technologies/Libraries Used

  • Statsmodels
  • Python
  • Matplotlib, Seaborn
  • Pandas, Numpy

Getting Started

  1. Clone this repository (for help see this tutorial).
  2. Raw Data is being kept [Raw Data](Repo folder containing raw data) within this repository.
  3. Data processing/transformation scripts are :
  • Task-1 Outlier Detection
  • Task-2 (pre-processing)
  1. Analysis is carried out in files names:
  • Task 2 Accounting Seasonality
  • Task 3 Comparing prices
  • Task 4 Flagset highest price Fluctuation
  1. Running the files in that order will do the trick.

Author:

Name Slack Handle
Vipul Rustagi LinkedIn

Screenshots

Raw Data

'1' '1'

Outliers within a commodity

'1'

Overall outliers

'1'

Price comparisons

'1'

Seasonality of each cluster of region and commodity pairs

'1'

Deseasonalised prices of Area:commodity after seasonal decomposition

'1'

Pairs with high price fluctuations

'1'

Contact

You can’t perform that action at this time.