dessskris Posted December 16, 2011 Report Share Posted December 16, 2011 Also, Im trying to put in values for the k l and m parameters, but i keep getting a straight horizantal line on my calculator (TI-83 graphing calc) and Im not sure why Im not getting a proper logistic curve I know what the variables are suppose to represent, i just cant seem to find the right values. Is it okay for me to ask if someone knows the ranges for the values,(e.g. K> B (?) ) ? that will make my curve, curve properly and not be a straight horizatal line just find the best regression by using your TI Hello. Just a quick question for "Analytically develop one model function that fits the data points on your graph.". If "Analytically" means algebraically, you cannot use your GDC and Wolfram Alpha at all? Or can I use my calculator to solve my simultaneous equation? because it seems really difficult to solve it only by hand. it just means you can't use your calculator to find the best regression. if you use matrices you can use technology to work out the matrices for you. so I've made my graph of the function that fits the data (it's exponential) and put it on the same axes as the original data. It's way off... it only passes through 1 point, and the last part says to revise your model if neccessary. Does that mean I need to make a new model? yes make a new model I have tried to "analytically" get a function for a cubic formula. The first time only like the first three point fit, so I tried to work it out with three other point. But I don't get a good model. I am using f(x)= ax3+bx2+cx+d Can someone spot a mistake in the function above? I also tried a exponential but also I am not getting a proper fit. Thanks for the help in advance. there's no mistake. have you converted the year values into small integers? Do we have to show mathematically how we interpreted the parameters for the researcher's mode? Or can I say I just used my calculator to come up with? can just say you got those values from your calculator, because you're asked to use technology anyway. BUT you need to show a screenshot. Reply Link to post Share on other sites More sharing options...
omgpop Posted December 18, 2011 Report Share Posted December 18, 2011 (edited) Unfortunately, I am confused...It says "What type of functions could model the behaviour of the graph?" I got sixth polynomial because I tried every trendline in Excel, and this one was the best, but somehow I feel that is not an appropriate explanation for IB... How would I explain my choice, as it asks for. Also, it says "choices" in the portfolio. Am I supposed to pick more than one trendline, or something? I really think it's sixth polynomial, though, since it gave me a superb trendline....I kind of wanted to justify my choices through a mix of just looking at the graph, but also commenting on the R squared value. Like for sixth polynomial, it's 0.99999999, which I know is good... Is that acceptable, or am I missing something? Edited December 19, 2011 by omgpop Reply Link to post Share on other sites More sharing options...
tank85 Posted December 20, 2011 Report Share Posted December 20, 2011 agh I feel so dumb I screwed up the last section of the portfolio. Will that affect my grade? Reply Link to post Share on other sites More sharing options...
omgpop Posted December 21, 2011 Report Share Posted December 21, 2011 (edited) OK, so on top of my last question regarding how should I justify my choice of graph, it asks to "revise your model if necessary". My model fits really well, with the exception of the final two points. Should I just be like "close enough" (in fancy terms, of course) and is the function I am supposed to be getting (This is before the researcher's function is introduced) a sixth polynomial function? If it's not, then my function is just incorrect.Also, what does "Discuss the implications of each of these models..." mean? Models are just the researcher's model and the model you derived, correct? And I can't see any implications...And (I know it's a lot of questions, but I wanna ask them all in one go), for commenting on how well the models fit the new data, I am assuming it is not necessary to graph anything, right. Also, the very last thing, modifying the model, I do not know what to do... Any changes I make that results in the function being more accurate with respect to the additional data causes it to be less accurate with the original data provided. Edited December 21, 2011 by omgpop Reply Link to post Share on other sites More sharing options...
eseb Posted December 25, 2011 Report Share Posted December 25, 2011 I have read through the first 13 pages of this discussion and have not found the right program in order to be able to graph a logistic function, can someone please help i have searched a lot but i just couldnt find the right program i have downloaded graph 4.3, geogebra, autograph, excel, and looked for a software from the internet but found nothing! Reply Link to post Share on other sites More sharing options...
dessskris Posted December 27, 2011 Report Share Posted December 27, 2011 I have read through the first 13 pages of this discussion and have not found the right program in order to be able to graph a logistic function, can someone please help i have searched a lot but i just couldnt find the right program i have downloaded graph 4.3, geogebra, autograph, excel, and looked for a software from the internet but found nothing!TI calculator. Reply Link to post Share on other sites More sharing options...
dessskris Posted December 27, 2011 Report Share Posted December 27, 2011 Unfortunately, I am confused... It says "What type of functions could model the behaviour of the graph?" I got sixth polynomial because I tried every trendline in Excel, and this one was the best, but somehow I feel that is not an appropriate explanation for IB... How would I explain my choice, as it asks for. Also, it says "choices" in the portfolio. Am I supposed to pick more than one trendline, or something? I really think it's sixth polynomial, though, since it gave me a superb trendline.... I kind of wanted to justify my choices through a mix of just looking at the graph, but also commenting on the R squared value. Like for sixth polynomial, it's 0.99999999, which I know is good... Is that acceptable, or am I missing something? you MAY mention more than one functions that you think will fit the data well, but you're only asked to analytically model ONE of those. think. think about the number of points. stationary points? characteristics of sixth polynomial? agh I feel so dumb I screwed up the last section of the portfolio. Will that affect my grade? unfortunately, yes of course that will. OK, so on top of my last question regarding how should I justify my choice of graph, it asks to "revise your model if necessary". My model fits really well, with the exception of the final two points. Should I just be like "close enough" (in fancy terms, of course) and is the function I am supposed to be getting (This is before the researcher's function is introduced) a sixth polynomial function? If it's not, then my function is just incorrect. there's no incorrect function in type 2 IAs. when modelling your own function, you need to make use of the first and last points (and some points in between). basically you first say whether it fits very well, well, or does not fit. then you talk about the trend, and points that do not fit (if any). Also, what does "Discuss the implications of each of these models..." mean? Models are just the researcher's model and the model you derived, correct? And I can't see any implications... that basically means: what will happen to China population in the future based on those two graphs? talk about each model separately. And (I know it's a lot of questions, but I wanna ask them all in one go), for commenting on how well the models fit the new data, I am assuming it is not necessary to graph anything, right. for that part, you choose ONE of the previously modelled functions. graph it on the same axes with the new data and comment on how well it fits. Also, the very last thing, modifying the model, I do not know what to do... Any changes I make that results in the function being more accurate with respect to the additional data causes it to be less accurate with the original data provided. translation/stretching. if desperate, model a new function. Reply Link to post Share on other sites More sharing options...
aaaaxxxx Posted December 27, 2011 Report Share Posted December 27, 2011 how do I prove it resembles a sigmoid curve?let's face it - it doesn't Reply Link to post Share on other sites More sharing options...
eseb Posted December 29, 2011 Report Share Posted December 29, 2011 how do I prove it resembles a sigmoid curve?let's face it - it doesn'tit does, after you have you choices, and narrow them down, the only choice you will be left with is a sigmoid curve. when you are narrowing your choices down, put in mind that you are dealing with a population, for example a sine curve is not going to work because a sine graph continues to go up and down and that is not going to happen when dealing with population, and there is a similar reason to every other curve, hence, the only appropriate curve is going to be a logistic cruve, Reply Link to post Share on other sites More sharing options...
patientflees Posted December 30, 2011 Report Share Posted December 30, 2011 Guys, I started with this portfolio like 2 days ago, defined all relevant variables and parameters and I have no clue what to do with matrices. I mean... for what are we doing matrices here and how should we complete its columns and rows? I really don't understand the thing. Reply Link to post Share on other sites More sharing options...
eseb Posted January 2, 2012 Report Share Posted January 2, 2012 i am stuck in setting up the matrices to solve the equations, can someone please help! Reply Link to post Share on other sites More sharing options...
jessika Posted January 5, 2012 Report Share Posted January 5, 2012 So I thought it would be easiest to find the best fitting regression on my calculator and then do it analytically and I found that a cubic regression fits the data really well but I CAN'T get the equation to come out right when i use matrices. Any tips on what data to put into my matrix so that the right a,b,c and d values come out. My portfolios due next week so I need help now Reply Link to post Share on other sites More sharing options...
eseb Posted January 6, 2012 Report Share Posted January 6, 2012 i am not sure whether you can put 4 variables in a matrix, i think it is up to 3. so i think the best way to do it is by using elimination and substitution to limit your variables to 3 variables, and then use matrices. Reply Link to post Share on other sites More sharing options...
Aarushi Bhardwaj Posted January 16, 2012 Report Share Posted January 16, 2012 Does anybody know what type of functions could model the behavior of the graph? .. Please help. My submission is due very soon! Reply Link to post Share on other sites More sharing options...
kimmy_k Posted January 17, 2012 Report Share Posted January 17, 2012 I am curious has to how this is supposed to be formatted? More specifically how the paper should be outlined, or if anyone could just give me any advice on how to start the writing portion? Reply Link to post Share on other sites More sharing options...
aneedanachos Posted January 19, 2012 Report Share Posted January 19, 2012 I also have this IA. And I'm am utterly confused as to how I can find the parameters for the first part when I am choosing a function. Please help! Reply Link to post Share on other sites More sharing options...
svesse Posted January 19, 2012 Report Share Posted January 19, 2012 When it says "Comment on any apparent trends shown in the graph. What types of functions could model the behavior of the graph? Explain your choices." How many graphs do you make with different best-fit lines?I have started with a linear, then I am going for the polynomial, is that enough or is it relevant to use more graphs between? Reply Link to post Share on other sites More sharing options...
omri Posted January 23, 2012 Report Share Posted January 23, 2012 Hey guys, ok here' a FAQ since there are so many of the same questions answered:What is a parameter?A parameter is, in understandable terms what a, b and c are in the quadratic function ax2+bx+c. They are constant for a specific equation, but not for every equation (just like quadratic equations). When it says define the parameters, just write down the constants you get when forming function equations.What graphing software should I use?Graph and Autograph have everything you need and are simple to use. Just type in the equation and it will graph it for you. You can also copy cells from Excel and it will plot a data series for you.How do I find K, L and M?As mentioned previously, use your calculator, press stat->edit-> plot your data. Then press stat->calc->logistic. And there you go.What do the values K, L and M mean?Try solving this algeberically.. plugging in when t=0, for example.. You'll figure this out.I also have a question, when it says analytically develop, do they really mean without technology? And what is this talk about matrices? 1 Reply Link to post Share on other sites More sharing options...
mavv Posted January 24, 2012 Report Share Posted January 24, 2012 Hey guys, ok here' a FAQ since there are so many of the same questions answered:What is a parameter?A parameter is, in understandable terms what a, b and c are in the quadratic function ax2+bx+c. They are constant for a specific equation, but not for every equation (just like quadratic equations). When it says define the parameters, just write down the constants you get when forming function equations.What graphing software should I use?Graph and Autograph have everything you need and are simple to use. Just type in the equation and it will graph it for you. You can also copy cells from Excel and it will plot a data series for you.How do I find K, L and M?As mentioned previously, use your calculator, press stat->edit-> plot your data. Then press stat->calc->logistic. And there you go.What do the values K, L and M mean?Try solving this algeberically.. plugging in when t=0, for example.. You'll figure this out.I also have a question, when it says analytically develop, do they really mean without technology? And what is this talk about matrices?analytically develop means using a mathematical method, try selecting a few points and setting up simultaneous equations for parameters, obviously if you have a number of parameters you will need a number of simultaneous equations. A good way to solve a set of simultaneous equations is with matrices. Reply Link to post Share on other sites More sharing options...
mavv Posted January 24, 2012 Report Share Posted January 24, 2012 When it says "Comment on any apparent trends shown in the graph. What types of functions could model the behavior of the graph? Explain your choices." How many graphs do you make with different best-fit lines?I have started with a linear, then I am going for the polynomial, is that enough or is it relevant to use more graphs between?provided you explain why you have chosen a polynomial and what consider what degree to use? Alternatives could be trigonometric, exponential etc.... 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.