Challenge 11 - Finding outliers, my take
...efine representative? The approach I provide below is my take on the problem. The algorithm below groups observations into like and dissimilar sets based on distances between the points. Unlike the s...
...efine representative? The approach I provide below is my take on the problem. The algorithm below groups observations into like and dissimilar sets based on distances between the points. Unlike the s...
Outlier screening is an anomaly detection methodology where the data is typically numeric. JMP has four outlier screening methods under the Analyze > Screening > Explore&nbs...
...ith ties. We could continue ranking them sequentially or give them the same value. The drawback to the former is that we may have the same value appearing in more than one group. The drawback to the l...
My apologies for the delay on this write up. I have been very busy with, among other things, preparations for a talk on Functional Data Explorer at JMP Discovery Summit . It appears the regular ...
...o map one set of values to a new set of values? For continuous data, it often makes sense to group the values into quantiles, deciles (i.e., 10 groups) possibly being most common. Write a f...
...olumn and Group by that column. The union, with duplicates removed, appears as the first column in the resultant table. To find an intersection, start by creating a one column table for each input s...
The need to create intersections, differences, and unions, comes up when we need to know values in common, exclusive, or unique to two or more data sets. For this month, create three functions: i...
What is the fastest way to turn the elements in a list of number lists into characters? This challenge is an extension of the question found here in the Community. I decided to test eight approaches:...
...raige_Hales' code posted in the previous blog entry, I tried implementing the above approach using pattern matching. Surprisingly, the results took five to six times longer. I’m not sure exactly why, but m...
...esponse to a community post. Craige (@Craige_Hales), despite having retired from JMP Development, remains very active here in the Community, answering user questions, maintaining his Uncharted blog, and c...
For Challenge 6, we’re going to try something a bit different. Rather than having me tell you what the challenge is, we’ll do it the other way around. To keep things focused, I’m going to limit the s...
Working with matrices is both convenient and fast. Many functions that work on a single value can also be applied to an entire matrix. In addition, functions such as V Max, V Mean, and V Quantile can...
Hopefully, you've had fun with Challenge 4. In the future, if you have an idea for something that would make a good Challenge, please let me know. I’m always looking for good material! C...
Thanks again to those who participated, and thanks for your patience with this late post. Everyone was able to complete the challenges. Both the largest numbers added (~2 million digits) and largest ...
Things have been busy the last few weeks, and I’m still pulling together a write up for the previous JMP Challenge. I hope to have something out by the end of this week or beginning of next. In the m...
In last month’s challenge, I had to avoid creating a situation where overflow error might cause additional complications. In fact, the algorithm for modular exponentiation in the Miller-Rabin based i...
Note: A challenge entry from @David_Burnham was submitted prior to the deadline but was quarantined by my email system. I only discovered it after the original blog post was published. I am u...
This month’s challenge is back to basics – finding prime numbers. I am interested in two functions: getPrimes(n,t) – where n is a positive integer and t can be one of two values. If it is 1 get...
Programming wisdom suggests that turning a cyclical programming task into a linear one, vectorization, improves computation time. But is this true? Under what conditions? By how much? Challenge 1 pro...
...s your preference, you need to let me know in your entry. Post your entries to the blog. I must receive your final entry by 11:59 PM Eastern Time on the day of the deadline. I will consider only o...