Random Chart Generator

FF,

I've seen at least two random number chart generators created in Excel, posted on a few forums, including FF. They might be from the same person, but they both contain data using HLC.

Does anybody here have a random chart generator that uses OHLC, and even better, displays that information using candles.

Apart from showing data as OHLC, the main thing I'd like to see is the ability to have the data not show any gaps. So just a continuous price data candle chart, using values created with a random number generator. Similar to the ones already in existence, with those minor changes added.

I know nothing of Excel, and would not know where to start in attempting to change the file I have included, to attempt to create what I'm looking for.

Thanks.

please let other people know what platform is?

anyway, after check the excel file, it is an picture.

Have modified to what I think you are asking for .

No guarantees as it is a long time since I played with XL

What is the purpose of this????

I will see if I can figure this out, I think you want an open price that is equal to the previous close in order to eliminate the gaps. Is that correct?

I too am curious why you want this, it appears the author intended to disprove the validity of technical analysis with this spreadsheet. Are you trying to use this to trade with?

If that is the issue then no problem ...

change formula in cell 'G7' (Coin Toss HLC Sheet) to '= J6' and copy down to 'G157'

HTH

I have modified the VBA code and chart style to suit your requirements. The attached file is an Excel 2007 file so, if that causes you a problem let me know.

Regards

Hi 1Day
You have done (haven't checked your Vb code) what I started with which is nearly there ... but if you check you will find that you have open prices higher than the high!!

I did amend that (above)... check it out and see what you think

Thanks for the replies everyone, the new versions are a definite improvement over the original.

I find the version created by 'futurespec' a little easier to use, with the large chart and candles.

Is there any way this version can have the change, as mentioned by J.W. Holmes where they say, 'I think you want an open price that is equal to the previous close in order to eliminate the gaps'.

So if one of the candles closed at 99.31, the next candle would open at 99.31.

My use for this is to see whether or not a certain 'pattern' exists in random data, or whether this pattern can only be found using real price data. This is why I'd like to remove the 'gaps' from the chart, so as to make it look like a typical chart. I understand this might remove some of the randomness, but I can't figure out another option, so it seems to be the best solution.

This should do it

HTH

it would be usefull if we can get more than 150 candles. something like 5000-8000 candles so that i can import into metatrader.

Nice Work!

Thanks I've been wanting to create this on my own, but this certainly helps to get me started.

I'm going to change a few things;
1) Create a mean/stdev value for the difference between OH/CH OL/CL, based on large sample historical data.
2) Adjust such that you can apply your own preffered number of 'tosses'.
3) Add basic indicators so that it looks more like a 'real' forex chart.

With ultimately the goal to provide the forex community with a poll on which they can vote 10 charts. 5 random charts and 5 real charts. And then vote which is a) for sure a real forex chart and b) which is for sure a random generated one.

I'd love to see the statistics of such a vote.

Why? Because I spent 340$ on Forex Tester 2 , which is basicly a way to visually backtest your charts. But what if I spent all my dollars, and future time, on graphs which could be considered Random (without fundamental/sentiment knowledge)?

But charts aren't random, far from it, they merely appear it at first.

The Excel random number generator isn't really random it's only pseudo random that is why you do seem to get momentum on it, if it was true random you'd get a straight line virtually.


Be warned!!

biggest load of crap i've read in awhile. go back to your journal where nobody has to read your patent nonsense. you understand.... nothing.... about anything... in the history of ever. hush and let the adults talk.

hoho, excel random number generator is NOT random. There are tricks to improve it, by convolution though but still could backfire.

that way i created 1m bars testing hh/hl stuff

I see Nubs doesnt understand the term pseudo random or now anything about computers surprise surprise.

Dont feed the trolls.


But do recompile mt4 ea's just to annoy him.

*slow clap for turveyd*

no i don't understand anything. i don't understand random or pseudo-random, and i don't understand how some random functions are seeded and how some random functions are less random than they otherwise could and should be, and i don't understand how some randomizing functions are completely flawed and broken due to bugs, and so on.

i don't understand anything. maybe if i study your journal real hard i might learn something... or maybe a cursory read through the entire thing for the last year or more is nothing but you pretending that 'next week you'll be making the big bux'. always next week, then next week comes and you are still tooling around going nowhere.

or maybe i've been writing code for the last 20+ years in a few different languages and am currently engaged in writing a compiler for a somewhat simplistic 8086-based assembler language. maybe i know more than you could imagine or understand.

and that's your problem turveyd. all that you think you know pales in comparison to some of the rest of us who generally quietly suffer your stupidity. you think you are an expert on a wide range of subjects, but you aren't even a novice at any of them. you struggle to string a coherent sentence together half of the time!

so please, keep enlightening us how excel is pseudo-random... as if that's something people don't know! keep enlightening us how a random series of numbers would produce a straight line. please, go on.

i suppose you have an excuse why the excel sheets if re-written in another language would still output similar results, right? you have an answer for everything and are always right, right? so what is it then? is it because they are also pseudo-random? what if after implementing the most demonstrably random function created so far this still produced the same types of results? what then genius? still too pseudo-random for you? what if someone took a coin and flipped it manually a few thousand times and still the results were similar? biased by the person flipping it i guess, right?

what exactly do you have to offer other than sound-bytes of pure deluded nonsense? i honestly wonder if you have some type of disability. the things you say are just uncanny.

this is about all the time i have to waste on someone so intellectually handicapped, so have at it turveyd. reveal to us your ongoing 'wit' and 'intelligence'. regale us in your wisdom.

i notice in at least one of the excel files it is signed-off by an associate professor of finance. hmm, let's listen to what turveyd has to say about whatever, instead of reading the explanation provided by a college professor! like all things, you miss the point and corrupt basic truths with your own version of reality. a sad, confused, broken reality.

that would be somewhat interesting. you could just programmatically generate data within mt4, have it spit out a bunch of random data as well as a segment of real chart-derived data to a csv file, and then you could see how often you yourself can pick the real data segment mixed-in amongst random data after having excel quickly generate a chart for you.

also, there are mt4 options for manually testing. flawed, but free is certainly cheaper than fxtester

Perhaps you should be asking yourself more fundamental questions, for example: what if the information I need to make a trade decision is not any of the data held in the OHLC of the bars for the timeframe I am looking at . . . what if the data I need is sometimes/often inside the bar . . .

Nubs mc troll has worked out how to do a google search!

Markets are not random the merely appear random, you can not beat a random market.