Backtesting in progress. APOGEE is coming.

backtest.py

@@ -0,0 +1,211 @@
+import json
+import csv
+
+print "Starting SIMULATION..."
+
+with open('quotes.json') as json_data:
+    data = json.load(json_data)
+    dates = []
+    for agroup in data:
+        for thedate in agroup.keys():
+            dates.append(thedate)
+    #get ordered dates
+    dates = sorted(dates, reverse=True)
+
+    #for sanitizing data. Yeah taking the longer route only for testing purposes.
+    di = []
+    for adate in dates:
+        for againgroup in data:
+            for againdate in againgroup.values():
+                if adate == againdate[0]:
+                    ki = {}
+                    ki["date"] = againdate[0]
+                    ki["second"] = againdate[1]
+                    ki["third"] = againdate[2]
+                    ki["fourth"] = againdate[3]
+                    di.append(ki)
+
+
+print "Loading PRICES..."
+
+with open('prices.csv') as csvfile:
+    readCSV = csv.reader(csvfile, delimiter=',')
+
+    ni = []
+
+    for row in readCSV:
+        for anelement in di:
+            if anelement["date"] == row[0]:
+                ki = {}
+                ki["date"] = anelement["date"]
+                ki["price"] = row[1]
+                ki["macd"] = anelement["second"]
+                ki["ema9"] = anelement["third"]
+                ni.append(ki)
+
+
+leli = ni
+
+num_day = 30
+
+results = []
+
+popo = 0
+
+for popo in range(1720):
+    ni = leli[(1720-popo):(1750-popo)]
+
+    this_date = leli[(1719-popo)]["date"]
+
+    prices = []
+    macds = []
+    ema9 = []
+
+    for i in range(num_day):
+        prices.append([float(ni[i]["price"]),ni[i]["date"]])
+        macds.append([float(ni[i]["macd"]),ni[i]["date"]])
+        ema9.append([float(ni[i]["ema9"]),ni[i]["date"]])
+
+    #checking and storing minimas
+
+    prices_minimas = []
+    macd_minimas = []
+
+    for k in range(len(prices)):
+        try:
+            if prices[k][0] < prices[k-1][0] and prices[k][0] < prices[k+1][0]:
+                prices_minimas.append(prices[k])
+            else:
+                pass
+        except:
+            pass
+
+
+    for t in range(len(macds)):
+        try:
+            if macds[t][0] < macds[t-1][0] and macds[t][0] < macds[t+1][0]:
+                macd_minimas.append(prices[t])
+            else:
+                pass
+        except:
+            pass
+
+
+    print "Price minimas are: " + str(prices_minimas)
+    print "MACD minimas are: " + str(macd_minimas)
+
+    buy_con = []
+    sel_con = []
+
+    macd_con_sell = True
+    macd_con_buy = True
+
+
+    for amin in range(len(prices_minimas)):
+        try:
+            if prices_minimas[amin] < prices_minimas[amin + 1]:
+                pass
+            else:
+                buy_con.append(0)
+                print "Price analysis failed for Buy condition."
+                macd_con_buy = False
+                break
+        except:
+            pass
+
+
+    for dmin in range(len(macd_minimas)):
+        try:
+            if macd_minimas[dmin] > macd_minimas[dmin + 1]:
+                pass
+            else:
+                buy_con.append(0)
+                print "MACD analysis failed for Buy condition"
+                macd_con_buy = False
+                break
+        except:
+            pass
+
+
+    for hmin in range(len(prices_minimas)):
+        try:
+            if prices_minimas[hmin] > prices_minimas[hmin + 1]:
+                pass
+            else:
+                sel_con.append(0)
+                print "Price analysis failed for Sell condition."
+                macd_con_sell = False
+                break
+        except:
+            pass
+
+
+    for nmin in range(len(macd_minimas)):
+        try:
+            if macd_minimas[nmin] < macd_minimas[nmin + 1]:
+                pass
+            else:
+                sel_con.append(0)
+                print "MACD analysis failed for Sell condition."
+                macd_con_sell = False
+                break
+        except:
+            pass
+
+
+    if float(ni[0]["macd"]) == float(ni[0]["ema9"]):
+        main_var = "CONTINUE"
+    if float(ni[0]["macd"]) < float(ni[0]["ema9"]):
+        main_var = "less"
+    if float(ni[0]["macd"]) > float(ni[0]["ema9"]):
+        main_var = "more"
+
+    cross_list = []
+
+
+    for aval in range(num_day):
+        #if float(aval["macd"]) == float(aval["ema9"])... This condition has to be developed for future.
+        if float(ni[aval]["macd"]) == float(ni[aval]["ema9"]):
+            print "OOPS! The unexpected happened!"
+            break
+        if float(ni[aval]["macd"]) < float(ni[aval]["ema9"]):
+            new_con = "less"
+        elif float(ni[aval]["macd"]) > float(ni[aval]["ema9"]):
+            new_con = "more"
+        if main_var == "more" and new_con == "less":
+            cross_list.append("cross_macd_now_less")
+            main_var = "less"
+            print "Cross Date: " + ni[aval]["date"]
+        else:
+            pass
+        if main_var == "less" and new_con == "more":
+            cross_list.append("cross_macd_now_high")
+            main_var = "more"
+            print "Cross Date: " + ni[aval]["date"]
+        else:
+            pass
+
+
+    if len(cross_list) == 0:
+        print "No cross occurs"
+    elif len(cross_list) == 1:
+        if cross_list[0] == "cross_macd_now_less":
+            macd_con_buy = False
+            print "You must SELL as per line-cross analysis."
+        if cross_list[0] == "cross_macd_now_high":
+            macd_con_sell = False
+            print "You must BUY as per line-cross analysis."
+    elif len(cross_list) > 1:
+        print "More than one cross. We will figure it out soon! :P"
+    else:
+        pass
+
+
+    print cross_list
+
+    if macd_con_buy:
+        results.append({"backtest_date": this_date, "test_name": "MACD", "result": "BUY", "operate_on": this_date})
+    if macd_con_sell:
+        results.append({"backtest_date": this_date, "test_name": "MACD", "result": "SELL", "operate_on": this_date})
+
+print results

bollinger_backtest.py

@@ -0,0 +1,134 @@
+import json
+import csv
+
+print "Starting SIMULATION..."
+
+with open('quotes.json') as json_data:
+    data = json.load(json_data)
+    dates = []
+    for agroup in data:
+        for thedate in agroup.keys():
+            dates.append(thedate)
+    #get ordered dates
+    dates = sorted(dates, reverse=True)
+
+    #for sanitizing data. Yeah taking the longer route only for testing purposes.
+    di = []
+    for adate in dates:
+        for againgroup in data:
+            for againdate in againgroup.values():
+                if adate == againdate[0]:
+                    ki = {}
+                    ki["date"] = againdate[0]
+                    ki["second"] = againdate[1]
+                    ki["third"] = againdate[2]
+                    ki["fourth"] = againdate[3]
+                    di.append(ki)
+
+
+print "Loading PRICES..."
+
+with open('prices.csv') as csvfile:
+    readCSV = csv.reader(csvfile, delimiter=',')
+
+    ni = []
+
+    for row in readCSV:
+        for anelement in di:
+            if anelement["date"] == row[0]:
+                ki = {}
+                ki["date"] = anelement["date"]
+                ki["price"] = row[1]
+                ki["macd"] = anelement["second"]
+                ki["ema9"] = anelement["third"]
+                ni.append(ki)
+
+
+leli = ni
+
+num_day = 30
+
+results = []
+
+popo = 0
+
+with open('quotes-bollinger.json') as json_data_bollinger:
+    data_bollinger = json.load(json_data_bollinger)
+
+    # for changing format of data. Yeah taking the longer route only for testing purposes.
+    di_bollinger = []
+    for adate in dates:
+        for againgroup in data_bollinger:
+            for againdate in againgroup.values():
+                if adate == againdate[0]:
+                    ki = {}
+                    ki["date"] = againdate[0]
+                    ki["upper_band"] = againdate[2]
+                    ki["middle_band"] = againdate[3]
+                    ki["lower_band"] = againdate[4]
+                    for aele in ni:
+                        avar = 0
+                        if aele["date"] == adate:
+                            ki["price"] = aele["price"]
+                    di_bollinger.append(ki)
+
+leli = di_bollinger
+
+for popo in range(1720):
+
+    this_date = leli[(1719 - popo)]["date"]
+
+    di_bollinger = leli[(1720 - popo):(1750 - popo)]
+
+    boll_array = []
+
+    for anele in range(num_day):
+        boll_array.append(di_bollinger[anele])
+
+    boll_array = [x for x in boll_array[::-1]]
+
+    # print boll_array
+
+    boll_list = []
+
+    for k in range(len(boll_array)):
+        try:
+            if boll_array[k]["price"] < boll_array[k + 1]["price"] and boll_array[k + 1]["price"] > boll_array[k + 2][
+                "price"]:
+                boll_list.append({"MAX": boll_array[k + 1]})
+            if boll_array[k]["price"] > boll_array[k + 1]["price"] and boll_array[k + 1]["price"] < boll_array[k + 2][
+                "price"]:
+                boll_list.append({"MIN": boll_array[k + 1]})
+        except:
+            continue
+
+    # print boll_list
+
+    for k in range(len(boll_list)):
+        try:
+            if boll_list[k].keys()[0] == "MAX" and boll_list[k + 1].keys()[0] == "MIN" and boll_list[k + 2].keys()[
+                0] == "MAX":
+                if float(boll_list[k].values()[0]["price"]) > float(boll_list[k + 2].values()[0]["price"]):
+                    print "Finding sell condition after date: " + boll_list[k + 1].values()[0]["date"]
+                    compare_value = float(boll_list[k + 1].values()[0]["price"])
+                    for athing in boll_list[k + 1::]:
+                        if float(athing.values()[0]["price"]) < compare_value:
+                            print "Sell condition on date: " + athing.values()[0]["date"]
+                            results.append({"backtest_date": this_date, "test_name": "MACD", "result": "SELL",
+                                            "operate_on": athing.values()[0]["date"]})
+                            break
+            if boll_list[k].keys()[0] == "MIN" and boll_list[k + 1].keys()[0] == "MAX" and boll_list[k + 2].keys()[
+                0] == "MIN":
+                if float(boll_list[k].values()[0]["price"]) < float(boll_list[k + 2].values()[0]["price"]):
+                    print "Finding buy condition after date: " + boll_list[k + 1].values()[0]["date"]
+                    compare_value = float(boll_list[k + 1].values()[0]["price"])
+                    for athing in boll_list[k + 1::]:
+                        if float(athing.values()[0]["price"]) > compare_value:
+                            print "Buy condition on date: " + athing.values()[0]["date"]
+                            results.append({"backtest_date": this_date, "test_name": "MACD", "result": "BUY",
+                                            "operate_on": athing.values()[0]["date"]})
+                            break
+        except:
+            pass
+
+print results