Need an algorithm to predict next number in a data set
Budget: $30 – $250 AUD
This project relates to horse racing, however no knowledge of horse racing is required. Please see the attached document. Here is an explanation:
Each row of the spreadsheet contains the details of a horse that has raced.
- In column K is the name of the horse. In column B is the date the horse raced.
- In the Blue column is a number. This is a numerical representation of the horses race performance on that day.
- In the green column I have created my own algorithm in which I tried to predict what that number was going to be, based on the previous instances of the number in the blue column.
- The red column is how accurate the prediction was.
For example, in row 20 I predicted that the horse named 'A A Agility' would produce the number 28.1. It produced 25.5 as seen in the blue column. The difference was 2.6. My prediction was based on the previous races for that specific horse (seen in rows 5-19).
The algorithm I need will produce a more accurate overall prediction than the one I have created. It is fine if the data needs to be analysed in a separate software, as long as it is able to be exported back into to excel once the prediction has been completed.
The full data set is around 30,000 rows of data.
Once the project is completed, I will need to be able to add to the data set and re-calculate the predictions on a regular basis.
Each row of the spreadsheet contains the details of a horse that has raced.
- In column K is the name of the horse. In column B is the date the horse raced.
- In the Blue column is a number. This is a numerical representation of the horses race performance on that day.
- In the green column I have created my own algorithm in which I tried to predict what that number was going to be, based on the previous instances of the number in the blue column.
- The red column is how accurate the prediction was.
For example, in row 20 I predicted that the horse named 'A A Agility' would produce the number 28.1. It produced 25.5 as seen in the blue column. The difference was 2.6. My prediction was based on the previous races for that specific horse (seen in rows 5-19).
The algorithm I need will produce a more accurate overall prediction than the one I have created. It is fine if the data needs to be analysed in a separate software, as long as it is able to be exported back into to excel once the prediction has been completed.
The full data set is around 30,000 rows of data.
Once the project is completed, I will need to be able to add to the data set and re-calculate the predictions on a regular basis.
Related categories:
Algorithm
Machine Learning (ML)
Mathematics
Statistical Analysis
Predictive Analytics