Determining Causality Between Two Data Sets

atcsharp

New member
Joined
Feb 24, 2016
Messages
1
Hey Everyone,

I am trying to determine causality between two data sets. I don't think that correlation is the best tool due to the nature of the data. I am looking for some help on what tool(s) will best accomplish my task.

My first data set is a cost where the values can be positive or negative. My second data set is a range of values that are always positive and fall within a range, say 2000-12000. The data sets are paired together in a time series. For example: the price was $20.50 and corresponding value was 5000.

What I would like to determine is when a price is extremely higher or lower, does it fall into a particular range indicating that price activity is from higher or lower values in the second data set. Thanks for the help and please let me know if I need to clarify anything.
 
Hey Everyone,

I am trying to determine causality between two data sets. I don't think that correlation is the best tool due to the nature of the data. I am looking for some help on what tool(s) will best accomplish my task.

My first data set is a cost where the values can be positive or negative. My second data set is a range of values that are always positive and fall within a range, say 2000-12000. The data sets are paired together in a time series. For example: the price was $20.50 and corresponding value was 5000.

What I would like to determine is when a price is extremely higher or lower, does it fall into a particular range indicating that price activity is from higher or lower values in the second data set. Thanks for the help and please let me know if I need to clarify anything.

What are your thoughts? What have you done so far? Please show us your work even if you feel that it is wrong so we may try to help you. You might also read
http://www.freemathhelp.com/forum/th...Before-Posting

Let's call data set one p (for price) and data set two v (for value) and use t for time. What I think you are saying is you have a set of data (ti, vi, pi) and want to know if there is some connection between the points. For example suppose you had the data points
t v p
0500020.5
12000-579.4
13000-379.4
2500020.7
26000220.7
3100001020.8
3120001420.8
4500020.9
4120001420.9
where say t was in months. Then the relationship would be
p = -979.5 + 0.1 * t + 0.2 * v
which, mathematically is a plane surface. Is this what you had in mind?
 
Last edited:
Top