# Spotting Patterns

#### mathisfunn

##### New member
Hey guys! I´ve been trying to spot a pattern in this data for some time now and I haven´t had any luck whatsoever. I was wondering if any of you fine individuals could take a look.
 1 4 3 2 3 5 3 1 2 4 4 5 5 2 4 6 4 2 7 4 4 8 1 3 9 4 3 10 5 2 11 3 2 12 3 5 13 3 2 14 2 6 15 6 5 16 0 4 17 4 5 18 4 2 19 4 6 20 2 2 21 7 3 22 8 0 23 4 4 24 4 2 25 3 3 26 8 4 27 2 2 28 5 4 29 5 5 30 0 3 31 6 5 32 2 1 33 7 5 34 4 1 35 6 3 36 6 2 37 1 3 38 4 6 39 1 2 40 5 7 41 2 4 42 4 6 43 2 6 44 4 4 45 0 7 46 4 2 47 1 9 48 2 1 49 6 4 50 4 4 51 2 5 52 3 6 53 3 0 54 5 6 55 3 2 56 2 6 57 4 8 58 2 4 59 7 1 60 1 3 61 5 5 62 3 3 63 3 5 64 1 7 65 7 1 66 5 3 67 2 6 68 6 1 69 0 5 70 6 4 71 3 2

Cheers!

#### ksdhart2

##### Senior Member
Well, I've run the numbers and found that your two data sets have a Pearson's Correlation Coefficient of -0.1815. This means there's a small negative correlation between them, and as one decreases the other increases. That's a small enough correlation that it's unlikely there's any pattern to be found. For reference, the sets of {x} and {x^2}, using 71 data points, have a Pearson's Correlation Coefficient of 0.9691, a very strong positive correlation.

If we plot the two data sets on an xy-plane, with the first set being the x-values and the second set being the y-values, then the linear regression line, sometimes called the "line of best fit," is y = -0.1778x + 4.4083. However, the y-values predicted by this line will differ from the actual values by approximately 1.8977. In general, given an x-value not present in the list, the predicted y-value will be only 3.29% accurate.

Additionally, I inputted both sets of numbers into the Online Encyclopedia of Integer Sequences and got no results for either set. All of this is not to say there is no pattern, just that it's unlikely.