0
$\begingroup$

I'm not sure if this is the right forum but here goes:

I have a requirement to attempt spot patterns in data , more specifically bank transaction data. There is transaction date, amount, description e.g.

6/2/2017 COFFEE 5$
6/2/2017 EMPLOYER DEPOSIT 1000$
10/2/2017 GROCERIES 50$
20/2/2017 EMPLOYER DEPOSIT 1005$
20/2/2017 GROCERIES 120$

There are hundreds of other entries for general transactions and credit card entries.

So in the above example would want to determine the employee deposit trend every two weeks.

I have extensive C#/general programming experience, just need some guidance as to what to look into. I understand this probably isn't a simple solution but any suggestions what areas to focus on would be appreciated.

If this isnt' the correct forum please let me know which one to go to.

$\endgroup$
1
  • $\begingroup$ Search this site for anomaly detection, there are many posts! $\endgroup$ Commented Feb 25, 2017 at 23:10

1 Answer 1

0
$\begingroup$

One of the options is to use ARIMA model that has implemented in many languages.

A very nice introduction is here (4 parts of reading, but worth to spend the time): http://ucanalytics.com/blogs/forecasting-time-series-analysis-manufacturing-case-study-example-part-1/

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.