Finding patterns in random data - Drops - overlappingPrev Topic Next Topic

New Topic New Poll

HeinRich-ZA

za
Member #254,013
July 21, 2025
10 Posts
Offline

Aug 7, 2025, 11:12 pm

The program looks for a sequence of numbers, n1, n2, ... mn, where n1< n2< ... mn
and then calculates the probability of each sequence.

E.g. If you have observed that the last 3 numbers were: n1< n2< n3
then the program will tell you the probability that the next number in the sequence will be n0.

You can also specify a minimum value for n0.
E.g. If you want to predict when a share price will drop low enough to make it worth buying.

For a free report, send a list of integers (no more than 5000) each on a separate line
Il2c

Everything

United States
Member #247,665
November 23, 2024
1,801 Posts
Offline

Aug 11, 2025, 11:47 pm

Elaborate.

61.4
86.5
0.16
what are the next 5 numbers?
HeinRich-ZA

za
Member #254,013
July 21, 2025
10 Posts
Offline

Aug 12, 2025, 10:54 pm

My programs look for long term trends. You question is irrelevant.
ddude003

Missouri
United States
Member #208,871
August 9, 2020
365 Posts
Offline

Aug 13, 2025, 9:10 pm

"E.g. If you want to predict when a share price will drop low enough to make it worth buying."

Are you predicting the financial markets or the Lottery?

Sounds like you are using some time series forecasting algorithms...
HeinRich-ZA

za
Member #254,013
July 21, 2025
10 Posts
Offline

Aug 13, 2025, 10:57 pm

My aim was the lottery. I begin to realize that my algorithms may be useful for any time-series analysis.

I am looking for someone who can provide me with some real time-series data (share prices, signal processing, etc) so I can test my algorithms.
ddude003

Missouri
United States
Member #208,871
August 9, 2020
365 Posts
Offline

Aug 14, 2025, 12:25 am

It is pretty easy to get/scrape financial and/or lottery time series data from many sources for analysis... Google is your is your frenemey... And then you might try something like finviz.com if your more inclined towards financial markets...

Good luck in your endeavors fren...
HeinRich-ZA

za
Member #254,013
July 21, 2025
10 Posts
Offline

Aug 14, 2025, 9:47 am

I have found, but it was not easy. I had to crawl through lots of pop-ups, badly designed websites, downloaded several useless formats, etc. It took 2 hours to 2 useful datasets.
Orange71

Texas
United States
Member #200,551
August 28, 2019
177 Posts
Offline

Aug 15, 2025, 11:21 am

Quote: Originally posted by HeinRich-ZA on Aug 7, 2025
The program looks for a sequence of numbers, n1, n2, ... mn, where n1< n2< ... mn
and then calculates the probability of each sequence.

E.g. If you have observed that the last 3 numbers were: n1< n2< n3
then the program will tell you the probability that the next number in the sequence will be n0.

You can also specify a minimum value for n0.
E.g. If you want to predict when a share price will drop low enough to make it worth buying.

For a free report, send a list of integers (no more than 5000) each on a separate line

If each number is unique, and there are m numbers, then the probability of a particular order (one direction, either least to greatest or vice versa) is 1/m!. If k (<m) numbers have already occurred in order, then the conditional probability the next number picked will be the next in sequence is 1/(m-k).
ddude003

Missouri
United States
Member #208,871
August 9, 2020
365 Posts
Offline

Aug 15, 2025, 8:49 pm

I have found that R has an abundance of time series analysis packages including financial time series data collection and analysis...

A sip from the fire hydrant https://cran.r-project.org/web/views/TimeSeries.html...

If I remember correctly there are some getSymbols type of function in the quantmod package...
ddude003

Missouri
United States
Member #208,871
August 9, 2020
365 Posts
Offline

Aug 15, 2025, 9:11 pm

Your favorite AI might even help you write the code to interactivity down load a financial instruments time series data, reformat to your liking and even run one or more time series analysis and forecasting algorithms...
HeinRich-ZA

za
Member #254,013
July 21, 2025
10 Posts
Offline

Aug 15, 2025, 11:17 pm

To find something in that fire hydrant that may be useful to me is a nightmare. Easier to write my own programs.
HeinRich-ZA

za
Member #254,013
July 21, 2025
10 Posts
Offline

Aug 15, 2025, 11:18 pm

I understand your maths, but my programs work differently.
ddude003

Missouri
United States
Member #208,871
August 9, 2020
365 Posts
Offline

Aug 16, 2025, 7:06 pm

Good luck in your efforts!!!
hypersoniq

Pennsylvania
United States
Member #1,340
April 6, 2003
2,939 Posts
Offline

Aug 17, 2025, 12:47 pm

Quote: Originally posted by ddude003 on Aug 15, 2025
I have found that R has an abundance of time series analysis packages including financial time series data collection and analysis...

A sip from the fire hydrant https://cran.r-project.org/web/views/TimeSeries.html...

If I remember correctly there are some getSymbols type of function in the quantmod package...

Good info!

Also of interest is this one

https://CRAN.R-project.org/view=DynamicVisualizations

Have an EXCELlent day!
ddude003

Missouri
United States
Member #208,871
August 9, 2020
365 Posts
Offline

Aug 17, 2025, 6:46 pm

A little cut and paste from one of my trend following hacks... Minus the trend following part... Leaving just the getting of data, reformatting & writing the reformatted data as a .csv file...

# # Get a financial symbols open, high, low, close & volume... # # library(TTR) library(quantmod) # today = Sys.Date() zdate = format(today, "%d %b %Y") # with month as a word # asym = getSymbols('GLD',from=today-65, auto.assign = FALSE) # open = Op(asym) colnames(open) <- c("Open") # high = Hi(asym) colnames(high) <- c("High") # low = Lo(asym) colnames(low) <- c("Low") # close = Cl(asym) colnames(close) <- c("Close") # vol = Vo(asym) colnames(vol) <- c("Vol") # # Bind the data together... # mydata <- cbind(open,high,low,close,vol) mydata <- as.data.frame(mydata) # # write the data as a Coma Seperated file... # write.csv(as.data.frame(mydata), file="gldTrend.csv") #

Oops formatting error... Should be pretty easy to reformat... Have fun...

New Topic New Poll

Subscribe to this topic

Finding patterns in random data - Drops - overlappingPrev TopicNext Topic

Finding patterns in random data - Drops - overlappingPrev Topic Next Topic