Mutual funds data for 10 years monthly frequency ( 10 types of schemes). How can we create a usage pattern or usage behaviour. Suggest any tool or pattern recognisition software which can be applied on metric data.
It would really help to have more details about your data. For example, Does it include prices or volume? Also, what do you mean by 10 types of schemes?
One approach we have used to detect patterns like you seem to be seeking is to use seasonal dummy variables and time as a variable. In your specific example, time as a variable might be numbered 0 through 119 and you might include dummy variables for up to 11 different months.
If you share more details about your data, I may have a better answer and I suspect others will share more ideas.
To the best of my knowledge, We have a vast area of research in the background of "Market Basket Analysis", which will be a part of your question. It needs major Data Analytics work to interpret the data incurred from such transactions.
I am having data of 10 types of MF schemes run by a particular Asset Management Company. The data has two components Volume ( Frequency) and Value of transaction ( Amount) over 10 years on monthly interval. I need to draw a usage pattern of user for the said AMC schemes. Please guide.
The buying behavior of the clients from the data can be patterned with the help of simple trend line and the periodic buying's of the clients individually and consolidated as a whole. For further pattern usage we can refer, " The Suggestion model " of Amazon and Flipkart. It includes certain algorithms to crack the patterns of the products.