class LINEARREG_SLOPE(CustomFactor):
'''
Slope of Trendline

Momentum Indicator  

**Default inputs** USEquityPricing.close  
**Default Window Length** 14  
'''  
inputs = [USEquityPricing.close]  
window_length = 15  

def compute(self, today, assets, out, close):  
    # Prepare X matrix (x_is - x_bar)  
    X = range(self.window_length)  
    X_bar = np.nanmean(X)  
    X_vector = X - X_bar  
    X_matrix = np.tile(X_vector,(len(close.T), 1)).T  

    # Prepare Y matrix  
    Y_bar = np.nanmean(close, axis =0)  
    Y_bars = np-tile(Y_bar, (self.window_length, 1))  
    Y_matrix = close - Y_bars  

    # Prepare variance of x  
    X_var = np.nanvar(X)  

    # multiply X matrix and Y matrix and sum (dot product)  
    # then divide by variance of X  
    # this gives MLE of beta  
    out[:] = (np.sum((X_matrix * Y_matrix), axis = 0)/ X_var) / (self.window_length)

def make_pipeline():
base_universe = Q500US
fast_slope = LINEARREG_SLOPE(inputs = [USEquityPricing.close], window_length = 15, mask = base_universe)
slow_slope = LINEARREG_SLOPE(inputs = [USEquityPricing.close], window_length = 30, mask = base_universe)

sma_5 = SimpleMovingAverage(inputs = [USEquityPricing.close], window_length = 5, mask = base_universe)  
sma_15 =SimpleMovingAverage(inputs = [USEquityPricing.close], window_length = 15, mask = base_universe)  

percent_change = (sma_5 - sma_15)/sma_15  

up_indicate = fast_slope>slow_slope  
down_indicate = fast_slope < slow_slope  

latest_price = [USEquityPricing.close.latest]  

longs = ((up_indicate) & (percent_change > 0.05))  
shorts = ((down_indicate) & (percent_change < -0.05))  

is_tradeable = (longs|shorts)  
return Pipeline(  
    columns = {  
        'longs':longs,  
        'shorts':shorts,  
        'percent_change':percent_change  
},  
    screen = is_tradeable  

)

results = run_pipeline(make_pipeline(), '2015-01-01', '2015-01-01')
results.head()