Tune PDQ value for ARIMA model using Python and MS SQL
Budget: $30 – $250 USD
I am looking for a Python and MS SQL expert to help me tune the PDQ value for an ARIMA model. (see details below)
Expected timeframe for completion: Less than a week
Skills and experience required:
- Strong proficiency in Python and MS SQL
- Experience with ARIMA modeling
- Knowledge of time series analysis and forecasting
- Ability to tune the PDQ value for optimal model performance
I have a fairly simple system for Technical analyze of the Nordic stock market. The technology is a Windows MS SQL server together with Python and .NET which collects daily data and performs a number of calculations. There is a forecast engine programmed using Python and has implemented the ARIMA model (time phased) which is using historical stock prices to predict the next 60 days stock price. The ARIMA model is using a P,D,Q value and uses the same P,D,Q value for all stocks. This I would like help to change.
I would like a freelance to write a new piece of code, to set the best PDQ value for a particular stock. There is a table which holds historical stock prices and the system is already prepared to accept a unique set PDQ value for a particular stock. (there is a field on stock level which should be populated). So this new python job, should, for each stock, read and analyse the historical data and determine the optimal PDQ for the particular stock, and update the PDQ field in the stock table. The code should also log in a table when it starts and ends. The final job shall be scheduled in Window scheduler and I anticipate it needs to be executed on a weekly basis over a weekend. So if a heavy processing is required, there is time for it to finish.
There are various ways to determine the PDQ value and in your response, please indicate your approach to set the PDQ value. The piece of code, should be written in Python and there is already Python code which you can re-use for accessing the MS SQL database.
The awarded freelancer, will be given access to a windows production environment for development and deployment. Look forward to your response and feedback.
Expected timeframe for completion: Less than a week
Skills and experience required:
- Strong proficiency in Python and MS SQL
- Experience with ARIMA modeling
- Knowledge of time series analysis and forecasting
- Ability to tune the PDQ value for optimal model performance
I have a fairly simple system for Technical analyze of the Nordic stock market. The technology is a Windows MS SQL server together with Python and .NET which collects daily data and performs a number of calculations. There is a forecast engine programmed using Python and has implemented the ARIMA model (time phased) which is using historical stock prices to predict the next 60 days stock price. The ARIMA model is using a P,D,Q value and uses the same P,D,Q value for all stocks. This I would like help to change.
I would like a freelance to write a new piece of code, to set the best PDQ value for a particular stock. There is a table which holds historical stock prices and the system is already prepared to accept a unique set PDQ value for a particular stock. (there is a field on stock level which should be populated). So this new python job, should, for each stock, read and analyse the historical data and determine the optimal PDQ for the particular stock, and update the PDQ field in the stock table. The code should also log in a table when it starts and ends. The final job shall be scheduled in Window scheduler and I anticipate it needs to be executed on a weekly basis over a weekend. So if a heavy processing is required, there is time for it to finish.
There are various ways to determine the PDQ value and in your response, please indicate your approach to set the PDQ value. The piece of code, should be written in Python and there is already Python code which you can re-use for accessing the MS SQL database.
The awarded freelancer, will be given access to a windows production environment for development and deployment. Look forward to your response and feedback.