
PreCalculus PreCalculus Math Forum 
 LinkBack  Thread Tools  Display Modes 
September 30th, 2018, 09:22 AM  #1 
Newbie Joined: Sep 2018 From: North America Posts: 1 Thanks: 0  Fitting Linear and Quadratic Functions to Data
Hello friends! I was hoping that you might be able to help me with these two questions that I have. I know it's a lot, so if you just want to help with a portion of one of the problems that is fine. All related comments are welcome. Thanks! 1. To determine the effect of alcohol consumption, a researcher has measured reaction times for seven volunteers who were given different amounts of alcohol. The data is given below. x = blood alcohol content (percent) y = reaction time (seconds) X Y 0.08 0.32 0.10 0.38 0.12 0.44 0.14 0.42 0.15 0.47 0.16 0.51 0.18 0.63 a) Make a scatter plot and comment on the strength of the linear relationship between the variables. b) Find the linear function that best fits this data. Graph the line on the same axes with your scatter plot. Is it a good fit? c) Use the model to predict the reaction times for an individual (not one of the above) whose blood alcohol content is 0.15 and 0.25. d) Do you think these predictions are accurate? Explain 2. Proper inflation is necessary to maximize tread life for car tires. Consider the following values for pressure and tire life for a certain model of tire. Pressure Tire life (lb/in^2) (1000s of miles) 26 50 28 66 31 78 35 81 38 74 42 70 45 59 a) Make a scatter plot of the data. Is there a linear relationship between pressure and tire life or would a quadratic model be a better fit? b) Find the quadratic function that best fits this data. Graph the function with your scatter plot. c) Predict the mileage for a tire inflated to 30 psi. d) Use the quadratic function to determine the optimum tire pressure to maximize tread life. 
October 1st, 2018, 07:59 PM  #2 
Senior Member Joined: Feb 2016 From: Australia Posts: 1,737 Thanks: 606 Math Focus: Yet to find out. 
One question at a time and provide your working or state clearly what you are having trouble with. This will improve you chances of receiving help and you'll probably learn something too!

October 1st, 2018, 10:41 PM  #3 
Global Moderator Joined: Dec 2006 Posts: 19,957 Thanks: 1844 
1. (a) On making the scatter plot, the points seem to lie approximately on two lines. The first three points lie exactly on the line Y = 3X + 0.08. The remaining points lie approximately on the line Y = 5.24X  0.318. Both of these are questionable models. There's too little data for my liking. (b) The phrase "best fits" is not defined. What definition are you supposed to use, and why? (c) As I've ducked part(b), I'll use my equation Y = 5.24X  0.318 to predict the points (0.15, 0.47) and (0.25, 0.99) (rounded to 2 decimal places). The "official" answers are probably nearer (0.15, 0.49) and (0.25, 0.77). (d) The given data is probably rather inaccurate and therefore not very suitable for making predictions, especially outside the range of the data. 2. (a) There certainly doesn't appear to be a linear relationship. A quadratic model would be somewhat better, but a cubic model would be a further improvement, and hence indicates that there's little reason to prefer a quadratic model (other than that a linear model looks very inappropriate). (b) Again, the term "best fit" needs to be defined. (c) They're probably expecting a prediction of 70, but I prefer 70.7. It isn't stated what model is to be used for the prediction. (d) 36 (lb/in²) 

Tags 
calculus, data, fitting, functions, linear, quadratic 
Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Non linear least curve fitting  Shamma  Math  0  January 6th, 2015 02:09 AM 
Multidimensional fitting of two sets of data  datameng  Linear Algebra  2  September 3rd, 2014 06:30 AM 
Are these really linear or quadratic functions?  PatyR  Algebra  5  February 9th, 2014 04:46 PM 
Two data sets fitting  Simon39  Abstract Algebra  4  November 29th, 2013 10:41 PM 
simple data fitting problem  dimper129  Linear Algebra  0  October 7th, 2009 08:01 AM 