BUGS FORUM: fun with math!
http://www.urlashare.com/forum/bugs/
Page 2 of 2 • Jump to page 1 2
#26
February 8, 2010 8:15 AM
calc.exe said:
***downloadhelper.net
cashinvested = 346 * 28 = 9688
totalvalue = 1910 * 28 = 53480
gpt = ((totalvalue - cashinvested) / cashinvested) * 100
gpt = ((53480 - 9688) / 9688) * 100
gpt = (43792 / 9688) * 100
gpt = 4.52023 * 100
gpt = 452.02
***thekingcenter.org
cashinvested= 8 * 183 = 1464
totalvalue = 46 * 183 = 8418
gpt = ((gpt + (((totalvalue - cashinvested) / cashinvested) * 100)) / 2)
gpt = ((452.02 + (((8418 - 1464) / 1464) * 100)) / 2)
gpt = ((452.02 + ((6954 / 1464) * 100)) / 2)
gpt = ((452.02 + (4.75 * 100)) / 2)
gpt = ((452.02 + 475.00) / 2)
gpt = (927.02 / 2)
gpt = 463.51
***martinlutherking.org
cashinvested= 5.5 * 183 = 1006.5
totalvalue = 49 * 183 = 8967
gpt = ((gpt + (((totalvalue - cashinvested) / cashinvested) * 100)) / 2)
gpt = ((463.51 + (((8967 - 1006.5) / 1006.5) * 100)) / 2)
gpt = ((463.51 + ((7960.5 / 1006.5) * 100)) / 2)
gpt = ((463.51 + (7.90909 * 100)) / 2)
gpt = ((463.51 + 790.90) / 2)
gpt = (1254.41 / 2)
gpt = 627.205
I'm so excited about this!!! Great work... Absolutely ready for production, But I think you can see there's quite a bit of math overhead here, and I don't like that one little bit, mostly because we can do better (and get better numbers to boot!! A real, full average!)
So, here's what I propose now that I've seen the calculation and have a better understanding of this whole system, let's chop this badboy down to do this in one loop: (Comments preceeded by ::X::)
::XX:: This is intended to "reset" the values of these variables if needed (you might be able ignore these next two lines.)
gpt=0
count=0
::X:: Start the loop to look at all values
startloop::
cashinvested= 5.5 * 183 = 1006.5
totalvalue = 49 * 183 = 8967
gpt = gpt + ((totalvalue - cashinvested) / cashinvested)
count= count + 1 ::X:: We're keeping this as a variable to later divide by, however, if you can get the total number of BUY/SELL paired transactions without needing this, you can also cut this out.
endloop:
::X:: Now we're out of the loop, next step is to turn GPT into a percentage [*100] and get the average out of the full GPT number [/count]
gpt=(gpt * 100)/count
-- And there it is- perfection in a smaller and lighter package.
#27
February 8, 2010 2:46 PM
#28
February 8, 2010 7:46 PM
NickSunshine said:
LMAO what? 
Holy sheet buhlz.. I have no idea what I was smoking earlier (hurrying to finish typing and get on with a meeting maybe??)-- Kinda funny though.. So, assuming you didn't fully understand my casually insane post from earlier....
No worries, everything is working. It's just another opportunity that I see to tighten up this code to produce even better results (mathematically and computationally) No need to over explain, we just get better results.
I came to this after admiring your new working code for a while and asking myself, "How could this be better?" as I do with every project I work on (you can imagine the hell I put my co-workers through, and how much the CEO likes my work
).
So, I figured out a simple way to reduce computational load, while improving the nature of the average calculation that is being done, basically there's just an overall better way to do this, while there is nothing wrong with the current setup. I just know from experience it's best to tighten up a project as much as possible before moving away from it.. Or something like that anyway.
So, the optimized code looks like this (we've added the variable count, it might not be needed because you might be able to easily tell the thing how many transaction pairs exist without counting like this):
codesnippet said:
gpt=0
count=0
startloop::
gpt = gpt + ((totalvalue - cashinvested) / cashinvested)
count= count + 1
endloop:
gpt=(gpt * 100)/count
The first part of this code resets variables (you might not need to do this)
The second part is the optimized loop (one calculation, no need for logic, and fewer calculations)
The third part is a conversion to get an average out of the variable gpt ...
------
Easy enough right? And it will make your servers happy (and keep any real mathematicians that might stumble on this at least somewhat pleased that we arrived at a reasonably clean and nice final result, no worries though, I'm sure they will have laughed at us until they've cried long before this point in our conversation..
)
#29
February 8, 2010 10:32 PM
sounds good hahahahhahah
hey we need a wide eyed smily like whoa?
hey we need a wide eyed smily like whoa?
#30
February 10, 2010 2:11 AM
Spurtis55 said:
hey we need a wide eyed smily like whoa?
#31
February 10, 2010 3:51 AM
cb123 said:
Holy sheet buhlz.. I have no idea what I was smoking earlier (hurrying to finish typing and get on with a meeting maybe??)-- Kinda funny though.. So, assuming you didn't fully understand my casually insane post from earlier....
LOL actually, i got most of it... just wanted to give you a hard time
cb123 said:
No worries, everything is working. It's just another opportunity that I see to tighten up this code to produce even better results (mathematically and computationally) No need to over explain, we just get better results.
I came to this after admiring your new working code for a while and asking myself, "How could this be better?" as I do with every project I work on (you can imagine the hell I put my co-workers through, and how much the CEO likes my work
).
So, I figured out a simple way to reduce computational load, while improving the nature of the average calculation that is being done, basically there's just an overall better way to do this, while there is nothing wrong with the current setup. I just know from experience it's best to tighten up a project as much as possible before moving away from it.. Or something like that anyway.
well, load isnt much of an issue, as the contents of the fugly script are only meant to be ran once... at that point it will write the GPT to the database, and future updates will occur at the point of sale, not much overhead at all there.
cb123 said:
So, the optimized code looks like this (we've added the variable count, it might not be needed because you might be able to easily tell the thing how many transaction pairs exist without counting like this):
codesnippet said:
gpt=0
count=0
startloop::
gpt = gpt ((totalvalue - cashinvested) / cashinvested)
count= count 1
endloop:
gpt=(gpt * 100)/count
The first part of this code resets variables (you might not need to do this)
The second part is the optimized loop (one calculation, no need for logic, and fewer calculations)
The third part is a conversion to get an average out of the variable gpt ...
------
yep, if we were doing it all every time, that would be cleaner...
now would be a good time to mention that this number will never be 100% accurate if we divide by two...
that assumes that you only bought once... however, the math looks at the average price per share over all purchases (ie: more shares became available, so i bought em)
we can bend the math in fugly and divide by that number if that makes sense (total buy transactions +1) ?
same for the first sale? it should be divided by something right?
cb123 said:
Easy enough right? And it will make your servers happy (and keep any real mathematicians that might stumble on this at least somewhat pleased that we arrived at a reasonably clean and nice final result, no worries though, I'm sure they will have laughed at us until they've cried long before this point in our conversation..
)
we are damn close... just need to hammer out the little details
#32
February 19, 2010 2:15 PM
#33
February 22, 2010 8:42 PM
#34
February 24, 2010 10:09 PM
#35
February 24, 2010 11:04 PM
LOL take your time!
i was just worried that you had given up or couldnt find page 2
i was just worried that you had given up or couldnt find page 2
#36
February 26, 2010 2:58 AM
updated fugly.php with 'the new math' -
all calculations:
all calculations:
fugly.php said:
GPT = ((GPT + (((TOTAL VALUE - CASH INVESTED) / CASH INVESTED) * 100)) / (PURCHASES + 1))Page 2 of 2 • Jump to page 1 2
YOU ARE NOT LOGGED IN.