Analytics: Signal Detection – Part 05 – Objective

I feel this is a good time to revisit previous signal detection challenges and move away from relatively simple ideas of measuring Pearson coefficient to determine relationship between time series – it’s too blasé at this point (after extensive coverage on this blog), though there is no staightforward way to do it using only Excel or other off-the-shelf tools. If this sounds novel to you, I recommend you revisit signal detection rubric before going on. Also, there is an interesting (somewhat technically convoluted) variation on Pearson correlation for signal detection which I will explore in future posts, but for now, let’s focus on something else.

Attached is a dataset I generated using a C++ algorithm (could have been written in C only.) This kind of problem isn’t very likely to arise in real business world setting (you’ll appreciate why when you have Eureka moment), but, I do genuinely believe, that it demonstrates the kind of intuition you should have to deal with enigmatic, mysterious (in some sense scientific) data challenges. If you can figure this one out quickly (less than 10 minutes), then you really are operating on a different level – a superior analyst.

As before, attached is an ASCII dataset with three columns: A, B, C. Column A denotes type of signal, column B sample number and C the actual signal value. In this problem, there is no time series per-se, i.e. sample numbers are just to enumerate datapoints, not to indicate any temporal relationship. Your task is to associate every signal name with only one other signal name (so in the end you’ll create pairs of signals.)

The clock is ticking. If you believe you’ve figured it out before 05/31/13, send me your solution to e-mail (see about page) and I will honor it $$ and post it here.

dataset_signal_detection_05.txt

Advertisements

~ by Monsi.Terdex on May 24, 2013.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
Normal Boy

Nothing out of the ordinary

Data Engineering Blog

Compare different philosophies, approaches and tools for Analytics.

%d bloggers like this: