Jump to content

Type II -- Population Trends in China


Karl H T

Recommended Posts

*sigh* just finished mine.

I like math IAs but I get frustrated quite easily when I'm stuck...

Especially the logistic regression. I couldn't find any computer software to do it so I spent a few hours on K/L/M alone..

Your GDC will produce a logistic regression, see earlier posts. Well done for finishing though
Link to post
Share on other sites

I was told to apply the logarithmic function to my model through the list option on the calculator where I get numbers for a, b, and c...what do those mean?:

Are you sure it was logarithmic and not logistic, a,b, and c are the parameters.
Link to post
Share on other sites

I was told to apply the logarithmic function to my model through the list option on the calculator where I get numbers for a, b, and c...what do those mean?:

The formula given to you on your calculator should look very similar to the researcher's model except with different variables (a, b, c instead of k, l, m).

Edited by holly
Link to post
Share on other sites

Can i get some help too please? Mines is due on thursday and I'm completely lost. How do you find a model?

If any one needs help with this task, I finished it last term, so just ask wink.gif

Got 19/20 points on it, so thats AWESOME biggrin.gif

Please i would really appreciate some/much help!!! dunno what its all about... at the moment i don't know where to start! any hints???

thanks

Link to post
Share on other sites

Can i get some help too please? Mines is due on thursday and I'm completely lost. How do you find a model?

It's not as difficult as it seems. First plot the points and see what "shape" it makes from the data the paper gives you. Let's say it looks like a parabola. Then your model might be a quadratic function (y = ax2+bx+c). Then plug in the x and y values you know (which is the first table they gave you) and find the missing variables (which in this case is a, b, and c) by using substitution. I actually used a very simple function to find my model so it's not as hard as you think. Good luck!

I hope this makes sense... I can explain more if you need.

Edited by holly
Link to post
Share on other sites

Guest Juicebox

basically, I was able to use matrices in order to find my equation. But how can I play around w/ that equation? I dont' really know how to do any of the analysis stuff, I basically just know how to find my models...

I used matrices to find a cubic equation but the GDC always says ERR:SINGULAR MAT has this happened to u?

Link to post
Share on other sites

If any one needs help with this task, I finished it last term, so just ask wink.gif

Got 19/20 points on it, so thats AWESOME biggrin.gif

I could use some help as well, what function did u use, i was thinking of using wither an exponential (so that the graph never goes below zero) or the cubic (cause if fits great with the points), but i have no friken idea how to find a,b,c,d (in a ax^3 + bx^2 + cx + d) or a,b (for ae^bx) analitically... any idea??

Anyway I could see your portfolio task?

Link to post
Share on other sites

Anyway I could see your portfolio task?

No, we won't do the task for you.

on the GDC when i select stat -> CALC -> Logistic I get values for a b and c how do I know which one is K L or M?

You can guess :P Or you can simply replace K,L and M by A,B,C and see when they accurately represent the data you were given.

In my IA, before going in to ABC, I analyzed and specified how the equation would react when the values of K/L/M increased/decreased, that way it gave me [more importantly, the reader] a better understanding of the Verhulst model.

basically, I was able to use matrices in order to find my equation. But how can I play around w/ that equation? I dont' really know how to do any of the analysis stuff, I basically just know how to find my models...

Your equations will most likely contain Y's, X's and constants. Try altering the constant values[coefficients] in front of the variables. If its a linear model [y = mx + b] then just play around with M and B and see how the equation is altered.

How would you revise a model? I got my cubic function but on the GDC, it only hits five of the ten points. I don't want to use the sixth polynomial thing on excel b/c I have no idea how to explain it. Any suggestions?

The sixth polynomial's "formula" isn't anymore complicated than a quadratic formula. When you say it "hits", how close are we talking? I mapped 6 different functions through Excel, and they all had a coefficient of determination of >0.95

What you have to keep in mind, is your trendline's "behaviour" over a longer period of time. Does it accurately represent China's past and future demographics?

I got 18/20 on my most recent math IA, I'm quite happy.

If you have any specific questions, feel free to post. But asking for an IA, or the model that fits best, is not available.

Most important lesson learned was: correlation does not imply causation.

Edited by Stanislaw
  • Like 1
Link to post
Share on other sites

Basically it means the limiting factors. Like can your x value be a negative number? Stuff like that.

@Stanislaw: Meaning that it touches the points. Basically when I graph the points with my line, six of the points are colored in on the GDC. And I tried finding the sixth polynomial function using matrices but I got a weird graph, like a tangent function. I don't know what I'm doing wrong b/c I doing the same thing I do with finding my cubic function: I use matrices for both, except all I'm doing is I'm adding more rows and columns. I used excel in order to plot the function, and it was perfect for the population from 1950 to 1995, so could i just use that? I don't know how to show mathematical process for it becuase I just put it in on excel and it wasn't working when I used matrices. So I'm not sure what to do. The logistic function is best when thinking about China's past and future demographics. But since the IA says that were supposed to investigate diferent model functinos that best model the population from 1950 to 1995, I'm thinking we're only supposed to focus on what we're given rather than overall

Link to post
Share on other sites

idk if this is against the rules or not so inform me if it is, but when i use sinreg on the data using a ti 84, it spits out sin(6.2x+3.1)+875, and when i graph it, it's just a line and not a sinusoidal. (the numbers are much longer than that but you get the point) but when i use graphmatica, it spits out a different set of numbers and it actually forms a sinusoidal.

Link to post
Share on other sites

idk if this is against the rules or not so inform me if it is, but when i use sinreg on the data using a ti 84, it spits out sin(6.2x+3.1)+875, and when i graph it, it's just a line and not a sinusoidal. (the numbers are much longer than that but you get the point) but when i use graphmatica, it spits out a different set of numbers and it actually forms a sinusoidal.

I didn't use sin regression, I would suggest either using cubic, quartic (they end up being almost the same) or a sixth polynomial function. You can do all of this on Excel.

I have a question: for teh last part of the IA, when it asks us to plot the line along with the new points, I plotted it, and nothing is showing up on my graph. Just my points are coming up but the line is no one to be seen. WOuld anyone mind helping/explaining this? And this is for both my sixth polynomial function adn my logistic function

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...