Get on top of large data sets

Large data sets (more than a few hundred points) may seem like a gift for data analysis, but the reality can be more complex than that.  Time Series Plots and Control Charts become illegible, and you can miss some helpful patterns if you only analyze the complete data set.

There are two ways you can speed up your time to results:

• Take regularly-spaced samples across the whole data set – this makes it easier to see what is happening (so long as you still retain a large enough sample size)
• Take an extract of the data by using a 100% sample from a limited time frame – this can help you to see short-term behavior that is invisible when you look at the full data set

The Data Sampling Toolkit allows you to do both of these things, quickly and easily without the fuss of manually setting up sampling plans.

Take a representative sample of your data

The Data Sampling Toolkit makes it easy to take regular samples from your data, at whatever frequency you choose.

You will see the same overall patterns that exist in the full data set, but it will be easier to see what’s going on because the graphs will be a lot less crowded

Your source can contain up to 50,000 rows of data and the sample, up to 1,000 rows.

Look at a segment of your data

This option enables you to study variation over a shorter period of time.

You may be able to see patterns that are not visible when looking at all the data points, because a shorter time window enables you to see greater detail.

In this example, the segment of approximately two months reveals that there is a buildup at the end of the month that cannot clearly be seen in the full data set.