added the run time in distributed mode

proportions_and_regression.py

+#project code
+#group 3
+#Run time for distributed mode with 10 executer is 2m39s
 from pyspark import SparkContext
 import pyspark.sql.functions as F
 from pyspark.sql.functions import col, round
@@ -77,7 +80,7 @@ df_arrival=df.select('FlightNum', 'ArrDelay', 'DepDelay')
 
 #arrival delay time analysis
 from pyspark.mllib.stat import Statistics
-df_arrival=ddf_arrival.withColumn('ArrDelay', df_arrival['ArrDelay'].cast(IntegerType()))
+df_arrival=df_arrival.withColumn('ArrDelay', df_arrival['ArrDelay'].cast(IntegerType()))
 df_arrival=df_arrival.withColumn('DepDelay', df_arrival['DepDelay'].cast(IntegerType()))
 correlation=df_arrival.stat.corr("ArrDelay", "DepDelay")
 print(str(correlation)) #the correlation is 0.86 which means that there is a positive strong correlation between departure delay and arrival delay