forecast using ARIMA/GARCH of Natural gas price in MATLAB -- 2
Budget: €30 – €250 EUR
For this project, I am looking for a professional to utilize MATLAB to perform a forecast on the spot natural gas price (TTF) using GARCH/ARIMA.
I will provide at a later stage the full historical data on the data, but for the time being the closing in yahoo can be used (https://finance.yahoo.com/quote/TTF%3DF/history?p=TTF%3DF).
For the analysis, I would like MATLAB 2018 (or older) to be used. I would require you to have your own access to Matlab.
The goal is to provide a forecast on the price with an horizon of 1 to 5 days, and the accuracy of the forecast associated to it.
I m expecting someone :
1.Define the Input: The first step is to define the variables that need to be included in the model.
2.Collect and Analyze Data: Collect from my database historical data on natural gas prices, as well as other relevant variables that might impact prices, such as weather, supply, demand. Analyze the data to identify trends, patterns, and outliers that need to be taken into account.
3.Model Selection: Choose the appropriate GARCH model to use for the forecasting task. This can involve testing different models and comparing their performance using statistical metrics such as AIC or BIC.
4.Parameter Estimation: Estimate the parameters of the GARCH model based on the historical data. This involves using statistical methods such as Maximum Likelihood Estimation (MLE) to determine the optimal values for the model parameters.
5.Model Validation: Validate the GARCH model using statistical tests such as the Jarque-Bera test, which tests the normality of the residuals. This ensures that the model is a good fit for the data and can provide accurate forecasts.
6.Forecasting: Use the GARCH model to generate forecasts of natural gas prices for the desired horizon. This involves using the estimated model parameters and inputting them into the model to generate point forecasts and confidence intervals.
7.Evaluation and Review: Evaluate the accuracy of the forecasts using statistical metrics such as Mean Absolute Error (MAE) and Mean Squared Error (MSE). Review the forecasts regularly to ensure that they remain accurate and relevant, and update the model parameters as needed across various data slice.
8.integrate/implement the model with my backtesting tool to analyse historical forecast
This is a high-level project and requires specific expertise. If you are an experienced data scientist/statistician, with an aptitude for developing models and forecasting, I invite you to apply for this project.
I will provide at a later stage the full historical data on the data, but for the time being the closing in yahoo can be used (https://finance.yahoo.com/quote/TTF%3DF/history?p=TTF%3DF).
For the analysis, I would like MATLAB 2018 (or older) to be used. I would require you to have your own access to Matlab.
The goal is to provide a forecast on the price with an horizon of 1 to 5 days, and the accuracy of the forecast associated to it.
I m expecting someone :
1.Define the Input: The first step is to define the variables that need to be included in the model.
2.Collect and Analyze Data: Collect from my database historical data on natural gas prices, as well as other relevant variables that might impact prices, such as weather, supply, demand. Analyze the data to identify trends, patterns, and outliers that need to be taken into account.
3.Model Selection: Choose the appropriate GARCH model to use for the forecasting task. This can involve testing different models and comparing their performance using statistical metrics such as AIC or BIC.
4.Parameter Estimation: Estimate the parameters of the GARCH model based on the historical data. This involves using statistical methods such as Maximum Likelihood Estimation (MLE) to determine the optimal values for the model parameters.
5.Model Validation: Validate the GARCH model using statistical tests such as the Jarque-Bera test, which tests the normality of the residuals. This ensures that the model is a good fit for the data and can provide accurate forecasts.
6.Forecasting: Use the GARCH model to generate forecasts of natural gas prices for the desired horizon. This involves using the estimated model parameters and inputting them into the model to generate point forecasts and confidence intervals.
7.Evaluation and Review: Evaluate the accuracy of the forecasts using statistical metrics such as Mean Absolute Error (MAE) and Mean Squared Error (MSE). Review the forecasts regularly to ensure that they remain accurate and relevant, and update the model parameters as needed across various data slice.
8.integrate/implement the model with my backtesting tool to analyse historical forecast
This is a high-level project and requires specific expertise. If you are an experienced data scientist/statistician, with an aptitude for developing models and forecasting, I invite you to apply for this project.