Focus Fanatics - View Single Post - 12 Focus Depreciation That Bad? Really? <DATA ANALYSIS INCLUDED>
View Single Post
Old 01-30-2013, 08:11 PM   #35
Join Date: Mar 2011
Fan#: 84424
Location: LIVONIA, MI
What I Drive: .

Posts: 266
FF Reputation: 3 Jdog913 Good Standing Member
Buy-Sell-Trade Rating: (0)
Originally Posted by kam327 View Post
Consider the data while developing your own thoughts on the subject or ignore it, those are your only two choices. I have no intention on spending any more time on it.
Here are my thoughts. You presented nicely, comparing variables relative to others. (Good job there!) but you didn't check your data set for skewness. This is an extremely confusing topic for people who are not familiar with statistics.

Here is few questions to ask yourself.

Is the data for each vehicle normally distributed meaning a bell curve?

Does the arithmetic mean (Average) equal or to close the median of each data set?

Have you calculated the Z-Score for high and low values of each data set and make sure the score falls in between -2 and 2?

Now the Z-Score checks for outliers in the data set. (Really large or small numbers) Outliers can add bias to the data set and take away from the true average. A few variables that I thought of for this data set that can cause outliers are high mileage vehicles, and individuals who ask too much or too little for there vehicle. This can also be seen in the division of mean by the median. The closer it is the number one then the better.

All of the statements above relates to probability. Probability is important because if the results are no repeatable then the conclusion that was drawn is not correct. Now I've crunched some numbers that you posted for Tampa, FL. I tried to explain this the best I can but if you have any questions please feel free to ask. I'll try to explain Mean, Median, Standard Deviation and Z-Score the best I can.

I conclude that the Focus, Cruze and Corolla all have skewed data sets. Meaning that the data set is not normally distributed. This also means that the depreciation value calculated for the Focus, Cruze and Corolla are using non- normally distributed data. Lets, look at this more deeply.

The normal distribution graphs use a range of car values and see how many cars fall in between that value. If the number is Z on the X-Axis this means that <= Z and >(Z-1). For example 15 on the X-Axis range is <=15 and >14. Same goes for any other number.

2012 Focus

Sample Size= 20
Max = $17,900
Min = $12,300
Mean = $15,670
Median = $15,650
Std. Deviation = 1.39
Z Score Min = -2.41
Z Score Max = 1.59

2012 Cruze

Sample Size= 20
Max = $20,000
Min = $14,500
Mean = $16,495
Median = $16,900
Std. Deviation = 1.58
Z Score Min = -1.26
Z Score Max = 2.21

2012 Corolla

Sample Size= 20
Max = $17,400
Min = $12,900
Mean = $15,7900
Median = $16,000
Std. Deviation = 1.06
Z Score Min = - 2.72
Z Score Max = 1.51

2012 Civic

Sample Size= 20
Max = $19,000
Min = $15,000
Mean = $17,065
Median = $17,000
Std. Deviation = 1.04
Z Score Min = -1.96
Z Score Max = 1.84

2012 Elantra

Sample Size= 20
Max = $17,000
Min = $15,000
Mean = $16,035
Median = $16,000
Std. Deviation = .576
Z Score Min = - 1.79
Z Score Max = 1.67

I recommend changing the sample size and taking the middle 20 numbers with respect to the median of the whole data set causing a more normally distributed curve. Then we can make conclusions of the data.

When this is done then we will most likely see,

* Z-Scores great than -2 and less than 2
* Means and Medians extremely close (Around 1 when divided)
* Bell Shape Curve

Originally Posted by MyThIc3LiTe View Post
This data is statistically valid and doesn't need deeper analysis. It's pretty obvious the conclusions you can draw.

Don't see what engineering has to do with this either.

Then I would have to guess you don't know what engineering is about. Industrial Engineers deal with statistics on a regular basis. Other engineering disciplines deal with statistical data analysis.

Care to explain how the data is valid?
Jdog913 is offline  
    Reply With Quote