old codebase

src/silk_basket/core.py

@@ -0,0 +1,14 @@
+import pandas as pd
+import os
+
+def calc_gdp_benchmark():
+
+
+
+    pass
+
+
+
+
+
+

src/silk_basket/example/Kyle_workbook_v3.py

@@ -0,0 +1,71 @@
+import os
+
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+from src.silk_peg.utils import transfer_date_to_datetime, calc_ma, calc_perc_row
+
+if __name__ == "__main__" :
+
+    data_path = 'C:\scrt_projs\silk_peg\src\silk_peg\example\data'
+
+    price_data_df = transfer_date_to_datetime(
+        pd.read_excel(os.path.join(data_path, 'price.xlsx'), header=1),
+        'Year', format='%YYYY%mm%dd')
+    gdp_data_df = transfer_date_to_datetime(
+        pd.read_excel(os.path.join(data_path, 'GDP.xlsx'), header=0),
+        'time', format='%YYYY%mm%dd').replace([' NaN '],0)
+    gdp_capita_data_df = transfer_date_to_datetime(
+        pd.read_excel(os.path.join(data_path, 'GDP_capita.xlsx'), header=0),
+        'time', format='%YYYY%mm%dd').replace([' NaN '],0)
+
+    price_change_data_df = price_data_df.pct_change()
+
+    currency_country_LUT= { 'USDUSD': 'USA', 'CNYUSD': 'China', 'EURUSD': 'EU', 'JPYUSD': 'Japan', 'GBPUSD': 'UK',
+                            'INRUSD': 'India', 'CADUSD' : 'Canada', 'KRWUSD':'S. Korea', 'RUBUSD': 'Russia', 'BRLUSD': 'Brazil',
+                            'AUDUSD': 'Australia', 'IDRUSD': 'Indonesia', 'CHFUSD': 'Switzerland', 'TRYUSD': 'Turkey',
+                            'SEKUSD': 'Sweden', 'NOKUSD': 'Norway', 'SGDUSD':'Singapore','GOLD':'GOLD', 'Bitcoin':'Bitcoin'}
+    country_currency_LUT = {v: k for k, v in currency_country_LUT.items()}
+
+    gdp_perc = calc_perc_row(gdp_data_df)
+    gdp_capita_perc = calc_perc_row(gdp_capita_data_df)
+
+    price_moving_average = calc_ma(price_data_df, window=30)
+
+    benchmark_devision = gdp_perc.drop('Total', axis=1)
+    benchmark_devision = benchmark_devision*0.95
+    benchmark_devision['GOLD'] = [2.5 for i in benchmark_devision.index]
+    benchmark_devision['Bitcoin'] = [2.5 for i in benchmark_devision.index]
+    benchmark_devision = benchmark_devision.loc[benchmark_devision.index.repeat(11)]
+    benchmark_devision = benchmark_devision.iloc[385:,:]
+
+    new_value = []
+    benchmark_price = []
+    for (timestamp_row, row), i_row in zip(benchmark_devision.iterrows(), range(len(benchmark_devision)) ):
+        if not i_row == 0:
+            old_row = new_df.iloc[i_row-1,:]
+        else:
+            old_row = pd.Series([1 for i in range(len(row))], index=row.index)
+            new_df = pd.DataFrame(columns=row.index)
+
+        country_columns = [country_currency_LUT[currency] for currency in row.index]
+        new_row = price_change_data_df[country_columns].iloc[i_row + 1,:].values * old_row/100
+        # new_value.append(old_value * np.nanmean((price_change_data_df[country_columns].iloc[i_row + 1,:].values * row.values)))
+
+        new_df = new_df.append(new_row, ignore_index=True)
+        benchmark_price.append(new_row.sum(skipna=True))
+
+
+    # benchmark = gdp_perc*price_data_df
+
+    fig, axs = plt.subplots()
+    # axs.plot(price_data_df['EURUSD']/price_data_df['EURUSD'].mean())
+    # axs.plot(price_data_df['JPYUSD']/price_data_df['JPYUSD'].mean())
+
+    # axs.plot(price_moving_average['EURUSD'].iloc[:-6].index, new_value)
+    axs.plot(price_moving_average['EURUSD'].iloc[:-6].index, benchmark_price)
+    plt.show()
+
+
+    print('finito')

src/silk_basket/model.py

@@ -0,0 +1,9 @@
+
+
+def get_benchmark(name):
+    # name can be: GDP, world_trade,
+
+    if name == 'gdp':
+        calc_gdp_benchmark()
+
+    return

src/silk_basket/utils.py

@@ -0,0 +1,35 @@
+import numpy as np
+import pandas as pd
+
+def calc_ma(df, window, column=None):
+
+    if not column:
+        df = df.rolling(window=window).mean()
+    else:
+        df = df.rolling(window=window)[column].mean()
+
+    return df
+
+def calc_price_change():
+    pass
+
+def calc_perc_row(df):
+    if not 'Total' in df.columns:
+        df['Total'] = df.sum(axis=1)
+
+    df = df.replace(np.nan, 0)
+
+    perc_df = df.div(df['Total'],axis=0) * 100
+
+    return perc_df
+
+def transfer_date_to_datetime(df, column, format=None, set_index=True):
+    """
+    format example: '%YYYY%mm%dd'
+    """
+    datetime_col = pd.to_datetime(df[column], format=format)
+    df.loc[:, column] = datetime_col
+    if set_index:
+        df.set_index(column, inplace=True)
+
+    return df