def batchProcessRtnToNormalizedPrice(sourceFile, targetFile):
data = loadJson(sourceFile)
priceData = [1]
dataLabel = "data"
for rtn in data[dataLabel]:
priceData.append(priceData[-1] * (1 + rtn))
data[dataLabel] = priceData
with open(targetFile, "w") as file:
json.dump(data, file, indent=4)
def processBatchFinalRtnAndDumpToJson(fileNameWithoutDotJson):
tmp = dict()
datas = loadJson(fileNameWithoutDotJson + SUFFIX_OF_PRICED_DATA)
for count, data in enumerate(datas):
prices = datas[data]
tmp[str(count)] = prices[-1] - prices[0]
with open(fileNameWithoutDotJson + SUFFIX_OF_FINAL_RTN, "w") as file:
json.dump(tmp, file, indent=INDENT)
def processBatchMDDAndDumpToJson(sourceFile, targetFile=False):
tmp = dict()
pricedDatas = loadJson(sourceFile + SUFFIX_OF_PRICED_DATA)
for count, key in enumerate(pricedDatas):
tmp[str(count)] = MDD(pricedDatas[key])
if targetFile == False:
targetFile = sourceFile
with open(targetFile + SUFFIX_OF_MDD_DATA, "w") as file:
json.dump(tmp, file, indent=INDENT)
def processBatchPriceToRtn(filePath, dumpFileName, dataRange):
'''
Input :
The raw data from Alpha Vintage
dataRange = [lowerBound, UpperBound]
Output:
"metadata": {
"flag":0,
"underlying": xxxx,
"lengthOfEachExperiment": None,
"numberOfExperiments": None,
"returnStyle":1,
"time":str(datetime.today())
},
"data": [
0,
1.2,
.
.
.
1.3
]
the first element in "data" satnds for the earliest data
'''
deserializedData = loadJson(filePath)
pathForUnderlying = ["Meta Data", "2. Symbol"]
underlying = findValueInDictByKeyName(pathForUnderlying, deserializedData)
header = createHeaderOfTemplateForJson(0, 1, dataRange, underlying)
pathForPrice = ["Time Series (Daily)"]
priceData = findValueInDictByKeyName(pathForPrice, deserializedData)
initPrice = 0
for dailyPriceData in priceData:
beforeUpperBound = convetStringToDate(dailyPriceData) <= dataRange[1]
afterLowerBound = convetStringToDate(dailyPriceData) >= dataRange[0]
if beforeUpperBound and afterLowerBound:
currentPrice = float(priceData[dailyPriceData]['4. close'])
rtn = (initPrice - currentPrice) / currentPrice
header["data"].insert(0, rtn)
initPrice = currentPrice
header["data"].pop(-1)
dumpDictToFile(header, dumpFileName)
def dumpUniformDistribution(numberOfSampling,
sourceFileName,
numberOfExperiments,
targetFileName=False,
dataFromMemory=False):
deserializedData = loadJson(sourceFileName)
pathForUnderlying = ["metadata", "underlying"]
pathForTimeRange = ['metadata', 'timeRange']
underlying = findValueInDictByKeyName(pathForUnderlying, deserializedData)
dataRange = findValueInDictByKeyName(pathForTimeRange, deserializedData)
print(dataRange)
dataToBeSerialized = createHeaderOfTemplateForJson(
flag=2,
returnStyle=1,
underlying=underlying,
dataRange=dataRange,
lengthOfEachExperiment=numberOfSampling,
numberOfExperiments=numberOfExperiments)
if not dataFromMemory:
with open(sourceFileName, "r") as fileOfData:
deserializedData = json.load(
fileOfData)["data"][-numberOfSampling:]
elif dataFromMemory:
deserializedData = dataFromMemory
for eachExperiment in range(numberOfExperiments):
listOfChosenIndex = randint(0,
numberOfSampling - 1,
size=numberOfSampling)
pathOfReturn = list()
for index in listOfChosenIndex:
pathOfReturn.append(deserializedData[index])
dataToBeSerialized["data"].append(pathOfReturn)
with open(targetFileName, "w") as targetFile:
json.dump(dataToBeSerialized, targetFile, indent=4)