def _loadNormalizedCloseValuesForStocks(self, exchangeName, dateFrom, dateTo):
nasdaqDataRetrieval = HistoricalDataRetrieval(exchangeName)
stockNames = nasdaqDataRetrieval.getListOfStocks()
normalizedStocks = []
timePrinter = RemainingTimePrinter("Loading " + exchangeName + " stocks data", len(stockNames))
for a in range(len(stockNames)):
timePrinter.printMessage(a)
stockName = stockNames[a]
stockArray = np.array(nasdaqDataRetrieval.loadAdjustedCloseValues(stockName, dateFrom, dateTo), np.float)
if stockArray.shape[0] > 100 and stockArray[-1]/stockArray[0] > 1.2:
normalizedStock = stockArray / stockArray[0]
normalizedStocks.append((stockName, normalizedStock))
timePrinter.printLastMessage()
return normalizedStocks
def _findBestSharpeRatioByBruteForce(self, stocksNormalizedCloseValues):
stocksCount = len(stocksNormalizedCloseValues)
timePrinter = RemainingTimePrinter("Searching for best sharpe ratio", stocksCount * stocksCount)
bestStockA = None
bestStockB = None
bestProportion = None
bestSharpe = -1000000
for a in range(stocksCount):
for b in range(stocksCount):
for proportion in self._stocksProportions:
timePrinter.printMessage(b + a * stocksCount)
if a != b:
stockA = stocksNormalizedCloseValues[a]
stockB = stocksNormalizedCloseValues[b]
if stockA[1].shape[0] == stockB[1].shape[0]:
sharpeRatio = statictics.calculateSharpeRatioForStock(stockA[1] + stockB[1]*proportion)
if sharpeRatio > bestSharpe:
bestStockA = stockA
bestStockB = stockB
bestProportion = proportion
bestSharpe = sharpeRatio
timePrinter.printLastMessage()
return bestStockA, bestStockB, bestProportion, bestSharpe
