ibtoohard Posted March 30, 2014 Report Share Posted March 30, 2014 Hi for my math exploration I'm hoping to find a function that accurately represents the growing population of China. So far I've fitted linear and an exponential curve onto my data and gotten the equation. However I believe that a logistic curve would be the ideal equation but I'm currently using Logger Pro and I have no idea how to fit a logistic curve onto my data. Any help would be very appreciated as this exploration is due very soon! Reply Link to post Share on other sites More sharing options...
Rahul Posted March 30, 2014 Report Share Posted March 30, 2014 (edited) Press the curve fit button, and then click "Define Function..." Then, input f(x)=C/(1+Ae^(Bx)) and Description: Logistic. Press "OK" and there you have it, a logistic curve fit! Edited March 30, 2014 by Rahul Reply Link to post Share on other sites More sharing options...
ibtoohard Posted March 30, 2014 Author Report Share Posted March 30, 2014 I tried that and it gives me a line like this so would you be able to point out what I did wrong? Reply Link to post Share on other sites More sharing options...
Rahul Posted March 30, 2014 Report Share Posted March 30, 2014 Try f(x)=C/(1+A*exp(B*x)) instead! I just tried that with a data set and I got a curve that works - let me know if this doesn't work for you. Reply Link to post Share on other sites More sharing options...
ibtoohard Posted March 30, 2014 Author Report Share Posted March 30, 2014 Hey man sorry about this but its still just giving me a straight line. I'm not sure if its my data points or what but it still doesn't work Reply Link to post Share on other sites More sharing options...
Rahul Posted March 30, 2014 Report Share Posted March 30, 2014 I'm not sure what's going on, then. I just tried it and got a curve of fit that worked. Perhaps it's your data points? I can't identify what the issue is otherwise, though, as I'm not sure. Could you send me your data set? Reply Link to post Share on other sites More sharing options...
ibtoohard Posted March 30, 2014 Author Report Share Posted March 30, 2014 Hopefully you can see this as a chart but here are my data points x y1700 2107150001710 2193320001720 2283010001730 2376370001740 2473550001750 1795385001760 1968380001770 2136132001780 2775544001790 3014871001800 2952733001810 3457172001820 3533777001830 3957847001840 4128148001850 4299310001860 3770000001870 3580000001880 3580000001890 3800000001900 4000000001910 4230000001920 4720000001930 4890000001940 5208430001950 5566130001960 6820240001970 8258120001980 9812000001990 11336825002000 12658300002010 1370536875 Reply Link to post Share on other sites More sharing options...
Rahul Posted March 30, 2014 Report Share Posted March 30, 2014 Unfortunately, it's not showing up as a chart for me. Do you think you could attach a word or excel doc with those points so I don't have to retype? Reply Link to post Share on other sites More sharing options...
ibtoohard Posted March 30, 2014 Author Report Share Posted March 30, 2014 Here you go! I put the points into an excel fileDATA POINTS MATH IA.xls Reply Link to post Share on other sites More sharing options...
Rahul Posted March 30, 2014 Report Share Posted March 30, 2014 Unfortunately, I wasn't able to get a logistic or an inverse logistic to fit. I did the best I could with a fit, though, and have put it into the file here. Sorry I couldn't help more! Reply Link to post Share on other sites More sharing options...
ibtoohard Posted March 30, 2014 Author Report Share Posted March 30, 2014 Thanks for all your effort! So would this be an inverse exponent fit? And does this mean that a logistic is impossible for my data? Reply Link to post Share on other sites More sharing options...
Rahul Posted March 30, 2014 Report Share Posted March 30, 2014 That's an inverse exponent fit, yes! You'd actually be looking for an inverse logistic fit, as the graph begins concave down and goes concave up but this inverse exponential seems to fit very well. It would likely be possible to fit an inverse logistic to this, but it would take a lot of legwork on your part, more than seems reasonable so I would be inclined to stick with this. Reply Link to post Share on other sites More sharing options...
ibtoohard Posted March 30, 2014 Author Report Share Posted March 30, 2014 Sorry about all these questions I'll try to make this my last one but if I already have an exponential fit in my math IA should I also put an inverse exponent fit? Sorry about this I can't really tell the difference between the two Math isn't really my strongest subject Reply Link to post Share on other sites More sharing options...
Rahul Posted March 30, 2014 Report Share Posted March 30, 2014 Not necessary, but if you want to go ahead. It shows more investment, that you did try other things. 1 Reply Link to post Share on other sites More sharing options...
Genri Posted April 7, 2016 Report Share Posted April 7, 2016 hey guys i am also working on my exploration for maths hl right now, and thank you Rahul, i am using your advise which is ver helpful, but i thought that isnt the logistics function like that: f(x)=(C/(1+A*exp(B*x)))+D Please correct me if i am wrong thank you Reply Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.