fix financial data

app.py

@@ -76,6 +76,57 @@ def getAddressInfo(streetName, number):
     except AddressNotFound:
         return None, None
 
+# Adds the financial data to the apartment data
+def populateApartmentData(aptDf):
+    print('Populating with financial data...')
+    gdpDf = pd.read_csv(f'./data/historicalGDP.csv', sep=';')
+    unemploymentDf = pd.read_csv(f'./data/historicalUnemployment.csv', sep=';')
+    interestRateDf = pd.read_csv(f'./data/historicalInterest.csv', sep=';')
+    gdpDf = interpolateTime(gdpDf)
+    unemploymentDf = interpolateTime(unemploymentDf)
+    interestRateDf = interpolateTime(interestRateDf)
+    aptDf['gdp'] = aptDf['soldDate'].apply(getValueFromTime, args=(gdpDf,))
+    aptDf['unemployment'] = aptDf['soldDate'].apply(getValueFromTime, args=(unemploymentDf,))
+    aptDf['interestRate'] = aptDf['soldDate'].apply(getValueFromTime, args=(interestRateDf,))
+    return aptDf
+    
+def interpolateTime(df):
+    df['date'] = pd.to_datetime(df['date'])
+    df = df.set_index('date')
+    df = df.resample('MS').mean()
+    df = df.interpolate(method='time')
+    return fixChange(df)
+
+def getValueFromTime(datetime, dataDf):
+    # Get the value from the dataDf at the given datetime
+    # If the datetime is not in the dataDf, print the datetime and return '0'
+    # First, set the day of the datetime to the first day of the month
+    datetime = datetime.replace(day=1)
+    try:
+        return dataDf.loc[datetime, 'value']
+    except KeyError:
+        # Try adding one month
+        nextMonth = datetime.month + 1
+        if nextMonth > 12:
+            datetime = datetime.replace(month=1)
+            datetime = datetime.replace(year=datetime.year + 1)
+
+def fixChange(df):
+    # Set change to be the difference between the current and previous price
+    df['change'] = df['value'].diff()
+    # If the change is Nan set it to 0
+    df['change'] = df['change'].fillna(0)
+    
+    return df
+
+def fixChange(df):
+    # Set change to be the difference between the current and previous price
+    df['change'] = df['value'].diff()
+    # If the change is Nan set it to 0
+    df['change'] = df['change'].fillna(0)
+    
+    return df
+
 def cleanAddress(x):
     # Remove "-" from the street
     x = ''.join(x.split('-'))
@@ -227,21 +278,22 @@ def sthlm(streetName, number, sqm, rooms, monthlyFee, monthlyCost, floor, yearBu
             columns=columnHeaders)
     
     for soldDate in dates.keys():
-        gdp, unemployment, interestRate = getFinancialInfo(soldDate)
+        # gdp, unemployment, interestRate = getFinancialInfo(soldDate)
 
         # Parse the input so we can run it through the model
         # Create a dataframe from the input values
         input_variables = input_variables.append(
             pd.DataFrame(
-                [[streetName,number,sqm,rooms,soldDate,monthlyFee,monthlyCost,floor,yearBuilt,brf,agency,lat,lon,gdp,unemployment,interestRate]], columns=columnHeaders))
-        
-    df = normalizeData(input_variables)
+                [[streetName,number,sqm,rooms,soldDate,monthlyFee,monthlyCost,floor,yearBuilt,brf,agency,lat,lon]], columns=columnHeaders))
+    
+    df = populateApartmentData(input_variables)  
+    df = normalizeData(df)
     df = xgbFix(df)
 
     pricePred = None
     if auto:
         # pricePred = autoPred(df)
-        None
+        '', 'Autogluon is not working right now, please try again later'
     else:
         pricePred = xgboostPred(df)
 
@@ -258,7 +310,7 @@ def sthlm(streetName, number, sqm, rooms, monthlyFee, monthlyCost, floor, yearBu
             if diff > 0:
                 result.append(f'If the apartment would be sold {explanation} it would have been worth more: {parsePrice(pred)} (+{parsePrice(diff)})')
             else:
-                result.append(f'If the apartment would be sold {explanation} it would have been worth less: {parsePrice(pred)} (-{parsePrice(diff)})')
+                result.append(f'If the apartment would be sold {explanation} it would have been worth less: {parsePrice(pred)} ({parsePrice(diff)})')
 
     return '\n'.join(result), ''
 

data/historicalGDP.csv

@@ -0,0 +1,15 @@
+date;value;change
+2025-01-01;631.64;1.70
+2024-01-01;629.94;1.00
+2023-01-01;628.94;-1.20
+2022-01-01;630.14;2.70
+2021-01-01;627.44;4.80
+2020-01-01;541.49;-2.94
+2019-01-01;533.88;1.99
+2018-01-01;555.46;1.95
+2017-01-01;541.02;2.57
+2016-01-01;515.65;2.07
+2015-01-01;505.10;4.49
+2014-01-01;581.96;2.66
+2013-01-01;586.84;1.19
+2012-01-01;552.48;-0.59

data/historicalInterest.csv

@@ -0,0 +1,104 @@
+date;value
+2025-12-23;2.84
+2025-11-23;2.84
+2025-10-23;2.84
+2025-09-23;2.84
+2025-08-23;2.84
+2025-07-23;2.84
+2025-06-23;2.84
+2025-05-23;2.84
+2025-04-23;2.84
+2025-03-23;2.84
+2025-02-23;2.84
+2025-01-23;2.84
+2024-12-23;2.84
+2024-11-23;2.84
+2024-10-23;2.84
+2024-09-23;2.84
+2024-08-23;2.84
+2024-07-23;2.84
+2024-06-23;2.84
+2024-05-23;2.84
+2024-04-23;2.84
+2024-03-23;2.84
+2024-02-23;2.84
+2024-01-23;2.84
+2023-12-23;2.84
+2023-11-23;2.84
+2023-10-23;2.84
+2023-09-23;2.64
+2023-08-23;2.64
+2023-07-23;2.64
+2023-06-23;2.64
+2023-05-23;2.64
+2023-04-23;2.64
+2023-03-23;2.64
+2023-02-23;2.64
+2023-01-23;2.64
+2022-12-26;2.64
+2022-11-30;2.50
+2022-09-21;1.75
+2022-07-06;0.75
+2022-06-08;0.25
+2022-05-04;0.25
+2022-02-16;0.00
+2021-12-01;0.00
+2021-09-22;0.00
+2021-07-07;0.00
+2021-04-28;0.00
+2021-02-17;0.00
+2020-12-02;0.00
+2020-09-23;0.00
+2020-07-08;0.00
+2020-04-29;0.00
+2020-03-18;0.00
+2020-02-19;0.00
+2020-01-08;0.00
+2019-10-30;-0.25
+2019-09-11;-0.25
+2019-07-10;-0.25
+2019-05-08;-0.25
+2019-02-20;-0.25
+2019-01-09;-0.25
+2018-10-31;-0.50
+2018-09-12;-0.50
+2018-07-04;-0.50
+2018-05-02;-0.50
+2018-02-21;-0.50
+2018-01-03;-0.50
+2017-11-01;-0.50
+2017-09-13;-0.50
+2017-07-05;-0.50
+2017-05-03;-0.50
+2017-02-22;-0.50
+2016-12-28;-0.50
+2016-11-02;-0.50
+2016-09-14;-0.50
+2016-07-13;-0.50
+2016-04-27;-0.50
+2016-02-17;-0.50
+2015-12-16;-0.35
+2015-11-04;-0.35
+2015-09-09;-0.35
+2015-07-08;-0.35
+2015-05-06;-0.25
+2015-03-25;-0.25
+2015-02-18;-0.10
+2014-12-17;0.00
+2014-10-29;0.00
+2014-09-10;0.25
+2014-07-09;0.25
+2014-04-16;0.75
+2014-02-19;0.75
+2013-12-18;0.75
+2013-10-30;1.00
+2013-09-11;1.00
+2013-07-10;1.00
+2013-04-24;1.00
+2013-02-20;1.00
+2012-12-19;1.00
+2012-10-31;1.25
+2012-09-12;1.25
+2012-07-11;1.50
+2012-04-25;1.50
+2012-02-22;1.50

data/historicalUnemployment.csv

@@ -0,0 +1,15 @@
+date;value;change
+2025-01-01;8.30;0.00
+2024-01-01;8.30;0.40
+2023-01-01;7.90;0.80
+2022-01-01;7.10;-1.56
+2021-01-01;8.66;0.37
+2020-01-01;8.29;1.46
+2019-01-01;6.83;0.47
+2018-01-01;6.36;-0.36
+2017-01-01;6.72;-0.27
+2016-01-01;6.99;-0.44
+2015-01-01;7.43;-0.52
+2014-01-01;7.95;-0.10
+2013-01-01;8.05;0.07
+2012-01-01;7.98;0.18