Exercise from lecture 3

Did the annual flow of the Nile river change after the Aswan dam was built?

## charge the table

nile <- read.csv("C:/Users/Claudio Alvarez/Desktop/nile.csv", header = TRUE)
nile
##     Annual Year Pre1905
## 1     1120 1871       1
## 2     1160 1872       1
## 3      963 1873       1
## 4     1210 1874       1
## 5     1160 1875       1
## 6     1160 1876       1
## 7      813 1877       1
## 8     1230 1878       1
## 9     1370 1879       1
## 10    1140 1880       1
## 11     995 1881       1
## 12     935 1882       1
## 13    1110 1883       1
## 14     994 1884       1
## 15    1020 1885       1
## 16     960 1886       1
## 17    1180 1887       1
## 18     799 1888       1
## 19     958 1889       1
## 20    1140 1890       1
## 21    1100 1891       1
## 22    1210 1892       1
## 23    1150 1893       1
## 24    1250 1894       1
## 25    1260 1895       1
## 26    1220 1896       1
## 27    1030 1897       1
## 28    1100 1898       1
## 29     774 1899       1
## 30     840 1900       1
## 31     874 1901       1
## 32     694 1902       0
## 33     940 1903       0
## 34     833 1904       0
## 35     701 1905       0
## 36     916 1906       0
## 37     692 1907       0
## 38    1020 1908       0
## 39    1050 1909       0
## 40     969 1910       0
## 41     831 1911       0
## 42     726 1912       0
## 43     456 1913       0
## 44     824 1914       0
## 45     702 1915       0
## 46    1120 1916       0
## 47    1100 1917       0
## 48     832 1918       0
## 49     764 1919       0
## 50     821 1920       0
## 51     768 1921       0
## 52     845 1922       0
## 53     864 1923       0
## 54     862 1924       0
## 55     698 1925       0
## 56     845 1926       0
## 57     744 1927       0
## 58     796 1928       0
## 59    1040 1929       0
## 60     759 1930       0
## 61     781 1931       0
## 62     865 1932       0
## 63     845 1933       0
## 64     944 1934       0
## 65     984 1935       0
## 66     897 1936       0
## 67     822 1937       0
## 68    1010 1938       0
## 69     771 1939       0
## 70     676 1940       0
## 71     649 1941       0
## 72     846 1942       0
## 73     812 1943       0
## 74     742 1944       0
## 75     801 1945       0
## 76    1040 1946       0
## 77     860 1947       0
## 78     874 1948       0
## 79     848 1949       0
## 80     890 1950       0
## 81     744 1951       0
## 82     749 1952       0
## 83     838 1953       0
## 84    1050 1954       0
## 85     918 1955       0
## 86     986 1956       0
## 87     797 1957       0
## 88     923 1958       0
## 89     975 1959       0
## 90     815 1960       0
## 91    1020 1961       0
## 92     906 1962       0
## 93     901 1963       0
## 94    1170 1964       0
## 95     912 1965       0
## 96     746 1966       0
## 97     919 1967       0
## 98     718 1968       0
## 99     714 1969       0
## 100    740 1970       0
## comparison of annual flow between 2 periods

aswanpre <- subset(nile, Pre1905 == 1)
aswanpost <- subset(nile, Pre1905 == 0)

summary(aswanpre)
##      Annual          Year         Pre1905 
##  Min.   : 774   Min.   :1871   Min.   :1  
##  1st Qu.: 962   1st Qu.:1878   1st Qu.:1  
##  Median :1110   Median :1886   Median :1  
##  Mean   :1072   Mean   :1886   Mean   :1  
##  3rd Qu.:1170   3rd Qu.:1894   3rd Qu.:1  
##  Max.   :1370   Max.   :1901   Max.   :1
summary(aswanpost)
##      Annual          Year         Pre1905 
##  Min.   : 456   Min.   :1902   Min.   :0  
##  1st Qu.: 759   1st Qu.:1919   1st Qu.:0  
##  Median : 845   Median :1936   Median :0  
##  Mean   : 851   Mean   :1936   Mean   :0  
##  3rd Qu.: 919   3rd Qu.:1953   3rd Qu.:0  
##  Max.   :1170   Max.   :1970   Max.   :0
## T-test
t.test(nile$Annual ~ nile$Pre1905, var.equal = T)  ## mean of two subset are not equal
## 
##  Two Sample t-test
## 
## data:  nile$Annual by nile$Pre1905 
## t = -7.557, df = 98, p-value = 2.211e-11
## alternative hypothesis: true difference in means is not equal to 0 
## 95 percent confidence interval:
##  -278.9 -162.9 
## sample estimates:
## mean in group 0 mean in group 1 
##           850.9          1071.8

## comparing variance
var.test(nile$Annual ~ nile$Pre1905)
## 
##  F test to compare two variances
## 
## data:  nile$Annual by nile$Pre1905 
## F = 0.7001, num df = 68, denom df = 30, p-value = 0.227
## alternative hypothesis: true ratio of variances is not equal to 1 
## 95 percent confidence interval:
##  0.3641 1.2501 
## sample estimates:
## ratio of variances 
##             0.7001

## wilcoxon test nos parametric

wilcox.test(nile$Annual ~ nile$Pre1905)  ## subset are not equal
## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  nile$Annual by nile$Pre1905 
## W = 304, p-value = 1.185e-08
## alternative hypothesis: true location shift is not equal to 0

the two periods compared are not equal