Scribd
Upload
6 views7 pages

Time Series Analysis Project

The project aims to select a profitable and stable stock portfolio using time series forecasting and statistical analysis, with data sourced from the yfinance API from 2020 to 2024. The final forecasting model used is SARIMA, and the selected stocks include ICICIBANK.NS, TCS.NS, and others, with weights assigned based on forecasted volatility. The project reflects on the challenges faced, including poor performance of certain stocks and suggests improvements like using an ensemble of models for better predictions.

Uploaded by

Deepak Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
6 views7 pages

Time Series Analysis Project

The project aims to select a profitable and stable stock portfolio using time series forecasting and statistical analysis, with data sourced from the yfinance API from 2020 to 2024. The final forecasting model used is SARIMA, and the selected stocks include ICICIBANK.NS, TCS.NS, and others, with weights assigned based on forecasted volatility. The project reflects on the challenges faced, including poor performance of certain stocks and suggests improvements like using an ensemble of models for better predictions.

Uploaded by

Deepak Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

TIME SERIES ANALYSIs

PROJECT

NAME-DEEPAK KUMAR
ROLL N0.-220123013
BRANCH-MATHEMATICS and COMPUTING

METHODOLOGY
 Objective: selecting a profitable and stable stock portfolio using
time series forecasting and statistical analysis.
 Data: Data source is yfinance API, time span is 2020-01-01 to
2024-12-31, and the data is sampled daily.
Decomposition: We use seasonal_decomposition function for each
stock to analyze trend, seasonality and residuals.
Forecasting Model:
 Models tried are ARIMA,SARIMA,FBProphet and Exponential
Smoothening.
 The orders for models are selected based on the ACF and PACF
plots of different stocks
Metrics of Comparison: Analyze models based on the mean RMSE,
MAPE and directional accuracy. We will select model with low RMSE
and low MAPE and high directional accuracy.

Stock Selection
 Volatility-based sizing: lower forecasted volatility = higher
weight. It reduces the risk in our approach. The volatility is
forecasted using GARCH(1,1).
 Correlation filtering: Explain your diversification approach
using Pearson correlation threshold for selecting independent
stocks.

FORECASTED RESULTS
The final model used for forecasting is SARIMA.
PORTFOLIO COMPOSITION

The final selected tickers are ICICIBANK.NS, TCS.NS, SBIN.NS,


TATAMOTORS.NS, DLF.NS, OFSS.NS, RELINFRA.NS, KPRMILL.NS.
The following are the volatility weights based on forecasted
volatilities. KPRMILL has very high volatility hence very low weight is
assigned to it.

The given below pie chart shows the composition of my portfolio.


PERFORMANCE ON STOCKGRO

The following is the performance of my stocks on stockgro.


Major losses: My major losses is due to KPRMILL and Tata
Mototrs stocks which perform very in comparison to my
prediction.
Best Stocks: Reliance infra and SBI were the best
performing stocks.

MODEL ACCURACY

The following are the metrics for comparison of stock price for
2 days. We can conclude that OFSS and KPRMILL did not
perform according to my model.

I used MAPE and RMSE as metrics of comparison.


The below shown data that there is sudden movement in
stock price of KPRMILL.
REFLECTIONS
WHAT WENT WRONG: Taking stocks which performed well
in previous week was a blunder by me.
The sudden shift in stock price of KPRMILL work against my
model.
IMPROVEMENTS:
 I should use ensemble of models for prediction that may
have worked better.
 I should have tune the orders of SARIMA model little bit
more.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy