Presentation on theme: "The data I chose to use for my project was the average age of death of people and how it has changed over the years. I though it would be interesting."— Presentation transcript:
The data I chose to use for my project was the average age of death of people and how it has changed over the years. I though it would be interesting to be able to predict how long people will live in the future. Looking at how the age that we die and trying to decipher causes for these changes can be fascinating. Predictions like these also depend on many factors that I will explain in more detail.
XY 197873.5 197973.9 198073.7 198174.1 198274.5 198374.6 198474.7 198574.7 198674.7 198774.9 198874.9 198975.1 199075.4 199175.5 199275.8 199375.5 199475.7 199575.8 199676.1 199775.5 199876.7 199976.7 200077 200177.2 200277.3 As shown to the left I collected my data from http://aging.senate.gov/crs/aging1.pdf and put it into a table. Also I put the data into my calculator by pressing STAT Edit then entering all the L1 and L2 data. http://aging.senate.gov/crs/aging1.pdf Then to create a stat plot I first mad sure that my first stat plot was turned on in my calculator, and then pressed GRAPH on my calculator. This created my stat plot.
Y= ax+b Y=.14161538x+73.499 R=.9722657372 In order to then calculate my linear regression I pressed STAT CALC 4: LinReg. To make sure that this data could be graphed I then pressed VARS Y-VARS Function Y1. Then by pressing enter my calculator showed me the equation (shown to left) of my regression. This also gave me my Correlation Coefficients.
Y=a*x^b Y= 72.70029x^.0153304 R=.9001759758 To calculate my power regression I again pressed STAT CALC PwrReg. Again to make sure this data was graphed I pressed VARS Y-VARS Function Y1. I then pressed enter and got the equation for the power regression and also the correlation coefficient.
Y=a*b^x Y=73.51575*1.00187^x R=.9728291689 BEST FIT FOR DATA! Finally to calculate my exponential regression I once again pressed STAT CALC ExpReg. To also graph this data I pressed VARS Y-VARS Function Y1. This puts the equation into my Y equals so that it can be shown on my scatter plot. Once I pressed enter not only did the calculator show my equation but also my correlation coefficient which allowed me to see that exponential regression would be the best fir for my data.
Exponential regression would be the best fit for the data because we are going to gradually be able to survive longer due to many factors. New technology allows people to survive terminal illnesses and live on life support. New medications allow people to live and survive even though sickness and natural aging. The more our technology develops and medicine continues to flourish the more out age of death will increase allowing us to live longer on average.
Year (inside data) My TestActual data 197873.653973.5 198774.909474.9 199576.043475.8 Year (outside data) My Test 200373.19454 201078.21606 201278.51400 When I tested the data within my set I got very accurate answers. My answers were very reasonably close to the actual data and therefore enforces even more the best fit if exponential regression. Also the predictions I made for the future are very accurate the age will increase but as it gets higher it is going to begin to slightly plain out because you can only live so long before medicine and technology dont become a factor. So my data is very valid and could be used to predict any dates in the future.
After gaining my data, creating a scatter plot, then using different regressions to find the one that best fits, I concluded that for the average age of death, exponential regression would be the best line of fit. When this data was tested it supported this idea also, within the data and in the future. Using these regressions we can predict what age we could live to in even the year 2050 and be fairly accurate. Using these simple regressions we can see into our future.