[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 94 KB, 472x332, statistics5.jpg [View same] [iqdb] [saucenao] [google]
4125050 No.4125050 [Reply] [Original]

morning statisticians

i've just finished uni and am now on holidays, so am working on a little project until the job market picks up early next year. More details about project in post following.

This however, is a general stats, data modelling, forecasting and optimisation thread. I find all topics fascinating and want something to read while coding. So professors, post grads, under grads and people in industry: get in here! What do you do? What are you working on? Are you enjoying what you are doing?

>> No.4125084

A statistician had his head in an oven and his feet in a freezer. When asked how he felt, he said "Eh, I feel fine."

>> No.4125090

>>4125084

LOL Actually, he would have said, "I'm a room temperature."

>> No.4125122

>>4125090
no

>> No.4125139

>>4125084
love stat jokes

Logic is a systematic method for getting the wrong conclusion with confidence.
Statistics is a systematic method for getting the wrong conclusion with 95% confidence.

>> No.4125143

i'll start. This project I'm working on is an idea that I got from someone on here some time ago. The idea is to gather data from the stock market at 5-10 minute intervals, then analyse that data to pick out global trends, trends within industries, cyclic patterns etc. But then, instead of using that information to try and provide buying advice, use it to generate word salad. So by assigning groups of words to different types of trends and occurrences, a quite weird assortment of words can be generated.

When the majority of stocks are on the rise, positively connotated words will appear. When stocks are falls, negative words will take their place. If a specific industry is taking heavy losses (eg: aviation industry in light of Qantas workforce strikes) then negative words that can be associated with that industry might appear.

I have a pretty good idea in my head of how I want to do all this. At the moment I'm just coding a small perl app (nearly done) that will gather all this data. Once it is done, I can set it to work for a week or two, and while it is doing that, I can begin coding the analysis side of things.

>> No.4125152

>>4125090
1/10

>> No.4125174

>>4125143
If all goes according to plan, I hope to extend the project further. If I create a similar app that scans popular news websites for common words and themes, I can then use those words to supplement the word salad that the stock market is creating. So for example, instead of just producing weird negative words when the stock market is crashing, catch-words from the news like GFC, home loans, commodities etc might pop up as well according to what the reason for the crash is.

I would probably restrict it to search only the business/commerce sections of news websites, but it could be interesting to see Kim Kardashian come up as a reason for stock market crashes.

>> No.4125219
File: 17 KB, 459x185, correlation.png [View same] [iqdb] [saucenao] [google]
4125219

Harr harr harr

>> No.4125227
File: 289 KB, 540x1498, significant.png [View same] [iqdb] [saucenao] [google]
4125227

>> No.4125231

>>4125219
bahahaha, I hadn't seen that one!

>> No.4125366

Another idea I had was to turn the word salad into funny little fantasy narratives. So instead of a jumble of words, actually slot them into constructs for randomised narratives. so if an industry is suffering due to a battle between two of the largest competitors, a little narrative is created with one CEO the knight and the other the black dragon. Can fill in lots of other little details according to other environmental factors.