I have number cluster which is including 53 elements. Their log2 equivalent is in this table...

I have number cluster which is including 53 elements. Their log2 equivalent is in this table. log2 equivalent increasing with row number of elements. How can i find next element #53. I will send 1 BTC to best idea! Thank you..

Other urls found in this thread:

sourcesale.com/cha.php
twitter.com/SFWRedditImages

>6k usd reward for trivial statistics problem

you're lying, fuck off

it is trivial for you but not for me. if you think it is easy. solve it in short time and send answer with your bitcoin address.

that much money tied to a stats problem can mean something fishy. im not risking being arrested (Veeky Forums reports illegal activity to the American Feds). what are you trying to crack? if I know its not nefarious I'll post the estimation algorithm you're in need of.

im not cracking anything. it is challange between me and someone. 1btc not too much for me. i dont want waste my time more. just send your algorithm and btc address.

Is there some reason why you don't just do a linear regression?

>linear regression is worth 1BTC
surely, OP must have tried this and needs higher accuracy

Either that or he's just a rich brainlet

>linear progression
>with log
Are you fucking retarded?

I would hope so, but it never hurts to ask.

How's precalc going, champ?

52.4165821456

are these in time series?

sorry 53... obviously

You're welcome, send the BTC to your local whorehouse

53.5189542 if they are in time series and the inverse log for it is around 12906561369293836.
Without knowing much about the variables there is no way for me to tell if its an accurate estimate. Usually if you have variables that aren't linearly correlated you can't do a linear regression. That's why you would log transform the data to get a linear relationship between the logs so you can perform a regression. Then you can use the linear model equation you obtain to predict your 53rd entry by plugging in 53. Hope i helped.
1DkaaNrWVxdDZaJbK4YTTq2ATZnA6BEWBX

you can use newton's method for interpolating polynomials which will give you a 53rd degree polynomial if you use all points. Using divided differences, you could feasibly do it by hand and I think some widely available software like matlab support the method.

is there an advantage to doing this over a linear regression?

it depends on what you're trying to do and why. If the numbers have a strong mathematical relation to each other (for example, they are values of some function with some error in measurements) then it would be accurately described by an interpolating polynomial. If they are variables affect by many real world factors, then regression would be a better predictive tool.

so uh, where's my bitcoin OP
1Bg3LS8NaTXUuYCwJcNWGv54CYeETwNZuZ

he was clearly lying about the money, i feel like that was clear after the first few posts

sorry but your answers wrong.i changed the challenge. now im giving all cluster with elements. find next #53 to #60 elements and get 50 BTC!

details : sourcesale.com/cha.php

Last word is wrong on my post. Reward is 5 BTC. find next #53 to #60 elements and GET 5 BTC!

Wait, not 5, but 1. I was just joking.

FIND #53 to #60 elements value and get 5BTC..

Suck my cock and get 10 free BTC!

you want the exact values or a way to best estimate them?

5 BTC for exact values but if you sure about estimated value we will review it and give feedback while calculate near values with your estimated value. if we can find true answer with your estimate of course we will your btc.

"i took algebra 2 freshman year. I'm smarter than you fuckers and know linear means y=mx+b"