The USDA provides data on the value of agricultural goods exported by each U.S. state from 2000 to 2015. The export value is measured in millions of dollars.
I observed the aggregate amount of agricultural goods exported by each state. A more detailed analysis by type of commodity might yield more insight into why certain states changed rankings over the course of 15 years.
These other categories are:
Animal Products: Beef, Pork, Hides, Other Livestock, Dairy, Broilers, Other Poultry
Plant Products: Veggies (Fresh), Veggies (Processed), Fruits (Fresh), Fruits (Processed), Tree Nuts, Rice, Wheat, Corn, Feeds, Grain Products, Soybeans, Soymeal, Vegetable Oils, Other Oilseed Products, Cotton, Tobacco, Other Plant Products
(I would also be interested in measuring a state’s geographic proximity to a NAFTA neighbor, using an indicator variable for if this state trades with a NAFTA neighbor, or considering the size of arable land in each state.)
agricultural.exports$BASE represents the vector values for 2010
agricultural.exports$YEAR15 represents the vector values for 2015
> weighted <- (agricultural.exports$`YEAR15`/1.38)
I wanted to analyze the 2015 values after they were adjusted for inflation. According to the U.S. Inflation Calculator website, $1.38 in 2015 equates to $1 in 2000. In other words, there was 38% inflation over these 15 years.
> boxplot(agricultural.exports$BASE, notch = TRUE, col = "blue", main = "M($) exports - 2000")
> boxplot(agricultural.exports$weighted, notch = TRUE, col = "red", main = "M($) exports - 2015 inflation adj.")
As we see, there are two exporting states that represent the outliers beyond the upper whisker of each boxplot. These states are California and Iowa. Illinois remains in third place over the 15 years, but is not considered an outlier, even though its exports an aggregate value a little less than that of Iowa.
Over the 15 years, the first quartile showed less variance. The same occurs in the upper whisker. This could maybe be described as a harmonization among some states.
It may be interested to group these states by regions or indicate whether they share a border with a NAFTA country. We would also group them as being inside or outside a certain distance from a NAFTA neighbor.
Not considering the agriculture export values as inflation adjusted can lead to form overly optimistic projects. We can see “exponential” growth in the exporting values of the top producing states. See the steam charts below for BASE (2000), YEAR15 (2015), and weighted (2015 adjusted for inflation).
1 | 233333447999
2 | 237
3 | 14
6 | 9
The decimal point is 3 digit(s) to the right of the |
2 | 018891556678
4 | 116
6 | 134
8 | 0
10 | 0
22 | 5
2 | 0013666677903
4 | 4668
6 | 2
16 | 3
The summary of agricultural exports gives us a nice breakdown of how the value of agricultural exports changed over 15 years. Alaska is the state that exported the least, measured in millions of dollars of agricultural product, whatever that product might be. Adjusting for inflation, Alaska doubled the value of its agricultural exports. California almost tripled the value of its agricultural exports. The value of agricultural exports has at least doubled in real terms for all 50 states.
STATE BASE YEAR15 weighted
Length:50 Min. : 6.204 Min. : 16.78 Min. : 12.16
Class :character 1st Qu.: 182.337 1st Qu.: 441.42 1st Qu.: 319.87
Mode :character Median : 698.612 Median : 1570.87 Median : 1138.31
Mean :1025.309 Mean : 2661.05 Mean : 1928.30
3rd Qu.:1327.824 3rd Qu.: 3589.07 3rd Qu.: 2600.77
Max. :6853.875 Max. :22546.99 Max. :16338.40
Now we can compare how some states “gain ground” over others.
|*The place movement is indicated for each state in relation to the place in occupied in the first column (year 2000).|
lm(formula = agricultural.exports$YEAR15 ~ agricultural.exports$BASE)
Min 1Q Median 3Q Max
-1779.2 -134.1 177.6 363.2 2375.7
Estimate Std. Error t value Pr(>|t|)
(Intercept) -419.19815 131.84596 -3.179 0.00258 **
agricultural.exports$BASE 3.00421 0.08474 35.454 < 2e-16 ***
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 701.3 on 48 degrees of freedom
Multiple R-squared: 0.9632, Adjusted R-squared: 0.9625
F-statistic: 1257 on 1 and 48 DF, p-value: < 2.2e-16
Even though some states switched positions gained ground over others during 15 years, the value of agricultural exports in 2000 is still a good predictor variable for the response variable, the value of agricultural exports in 2015.