def handler(grouped):
se = grouped.set_index('saledate')['SalePrice'].sort_index()
# se is the (ordered) time series of sales restricted to a single basket
# we can now create a dataframe by combining different metrics
conc = concat(
{
'MeanToDate': expanding_mean(se).shift(1).fillna(method='ffill'), # cumulative mean
'MedianToDate': expanding_median(se).shift(1).fillna(method='ffill'), # cumulative mean
'MaxToDate': se.cummax().shift(1).fillna(method='ffill'), # cumulative max
'MinToDate': se.cummin().shift(1).fillna(method='ffill'), # cumulative max
'PrevSale': se.shift(1).fillna(method='ffill'), # previous sale
'SaleCount': expanding_count(se) # cumulative count
},
axis=1
)
# bring back SalesID, needed for join
se = grouped.set_index('saledate')['SalesID'].sort_index()
conc['SalesID'] = se
return conc
def test_expanding_count(self):
result = mom.expanding_count(self.series)
assert_almost_equal(result,
mom.rolling_count(self.series, len(self.series)))
def test_expanding_count(self):
result = mom.expanding_count(self.series)
assert_almost_equal(result, mom.rolling_count(self.series,
len(self.series)))