
Real Analysis Real Analysis Math Forum 
 LinkBack  Thread Tools  Display Modes 
May 1st, 2011, 02:16 PM  #1 
Newbie Joined: May 2011 Posts: 6 Thanks: 0  Need help to found a formula for a data serial
Hello first time here, and to be honest i m here because i need help. I have a data serie : 100 0.12 200 0.25 300 0.42 400 0.68 500 1.08 600 1.5 700 1.83 800 2.12 900 2.51 1000 3.09 1100 3.72 1200 4.2 1300 4.64 1400 5.21 1500 6.07 1600 6.99 1700 7.71 1800 8.35 1900 9.21 2000 10.48 2100 11.85 2200 12.92 2300 13.87 2400 15.13 2500 17.02 2750 21.31 3000 26.69 3250 33.07 3500 41.03 3750 50.51 4000 61.96 4250 75.71 4500 92.18 4750 111.9 5000 135.46 5250 163.58 5500 197.06 5750 236.93 6000 280.42 6250 334.76 6500 396.85 6750 467.43 7000 547.32 7250 637.31 7500 738.28 7750 873.75 8000 1004.9 8250 1142.96 8500 1289.66 8750 1442.97 9000 1604.53 9250 1772.41 9500 1948.16 9750 2129.93 10000 2319.2 10250 2514.19 10500 2716.29 10750 2923.82 11000 3138.08 11250 3357.48 11500 3583.23 11750 3813.81 12000 4050.37 12250 4291.47 12500 4538.16 12750 4789.1 13000 5045.25 13250 5305.34 13500 5570.27 13750 5838.85 14000 6111.88 14250 6388.27 14500 6668.74 14750 6952.25 i ploted the numbers in excel, i created a trendline but now i m not able to found a general formula for this serial. Excel give me a function like: y = 6E21x6  3E16x5 + 4E12x4  2E08x3 + 6E05x2  0.0543x + 12.34 but it don t works. If someone can help me about....... Dc 
May 2nd, 2011, 06:51 AM  #2 
Newbie Joined: May 2011 Posts: 6 Thanks: 0  Re: Need help to found a formula for a data serial
any help?

May 3rd, 2011, 02:29 PM  #3 
Member Joined: Jul 2009 From: New Jersey Posts: 65 Thanks: 0  Re: Need help to found a formula for a data serial
Try doing a spline

May 3rd, 2011, 04:31 PM  #4 
Newbie Joined: Apr 2011 Posts: 5 Thanks: 0  Re: Need help to found a formula for a data serial
Hi dendricala, I did a quadratic regression on my TI84+ calculator with every fifth of your data points. The calculator gave me the following function: y=0,000050749347026081x^2+0,29690681072548x+271,2427198859 This is pretty accurate (R^2=0,9976). Excel only has limited regression possibilities as far as I know. I am also doing regressions for my bachelor thesis in micro economics. Here is a link to the excel file with your data (grouped), a graph (plot) of the data and the function i found. LINKY: http://www.2shared.com/file/qXz2PtD/Qu ... thfor.html Best regards thenarfer (from Norway) 
May 5th, 2011, 08:37 PM  #5 
Newbie Joined: May 2011 Posts: 6 Thanks: 0  Re: Need help to found a formula for a data serial
hello a lot. I did a try, but i m sure i m very impared also to copy a formula...... let consider the first couple of data: x y 100 0.12 i need to change x with 100 in your equation =(0,000050749347026081*100)^20,29690681072548*100+271,2427198859 the result is 248.2736775 i m sure i m doing somenthing wrog !!!! 
May 8th, 2011, 08:27 PM  #6 
Newbie Joined: May 2011 Posts: 6 Thanks: 0  Re: Need help to found a formula for a data serial
it s against the rule bump my post????

May 8th, 2011, 09:54 PM  #7 
Senior Member Joined: Jun 2010 Posts: 618 Thanks: 0  Re: Need help to found a formula for a data serial
dendricala, If you're going to bump your post, you have to be a bit more subtle about it, or else add some spice to it. Maybe try something really suave, like saying "I'm sorry, I think the 27th line in my long list of number should read 3000 26.66, not 3000 26.69, really sorry about that!" I think that would be more effective in getting people's attention. Ormkärr 
May 9th, 2011, 09:15 AM  #8 
Newbie Joined: May 2011 Posts: 6 Thanks: 0  Re: Need help to found a formula for a data serial
opps you right i apology 
May 12th, 2011, 04:14 AM  #9 
Newbie Joined: May 2011 Posts: 6 Thanks: 0  Re: Need help to found a formula for a data serial
look like i posted a real difficult problem.

May 12th, 2011, 10:58 AM  #10 
Senior Member Joined: Jun 2010 Posts: 618 Thanks: 0  Re: Need help to found a formula for a data serial
dendricala, Did the work and suggestions of the previous posters not help? I don't really understand what you are after here. Do you want a polynomial "best fit" or some other regression, or an equation which gives exactly your values at the prescribed points? You will probably have to compromise one way or the othereither get a wellfitting curve, or else pass through all your points with possibly horrible intermediate behavior. Did you look at thenarfer's link? Ormkärr 

Tags 
data, formula, found, serial 
Search tags for this page 
Click on a term to search for related topics.

Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Data set with result but no formula.  mHelgeland  Algebra  1  September 22nd, 2013 06:21 PM 
My found integral vs Wolfram Alpha's found integral  Obsessed_Math  Calculus  4  February 9th, 2012 03:18 PM 
A site claims to have found a formula for Ramsey's Theorem  Maxime  Applied Math  1  September 5th, 2010 05:20 PM 
Looking for a future performance index formula on a Data set  dunrented  Advanced Statistics  4  April 9th, 2009 06:58 AM 
I found a recursive formula for the nth prime!  brunojo  New Users  1  November 30th, 2007 11:59 AM 