多元数据序列的多元时间序列分析

Question

I am following this guide on my data. https://machinelearningmastery.com/multivariate-time-series-forecasting-lstms-keras/

I have data from the World Bank, 15 features from 1990 to today, but I have this time series from multiple countries. The above guide works when you have a large time series. How can I "compile" data from different countries while they still have the same time and where shall I look?

Best

import wbdata #pip install wbdata
indicators1 = {"EN.CLC.MDAT.ZS": "Droughts, floods, extreme temperatures (% of population, average 1990-2009)",
          "EN.ATM.CO2E.PP.GD":"CO2 emissions (kg per 2011 PPP $ of GDP)",
          "NY.GDP.PCAP.PP.KD": "GDP",
          "SP.POP.TOTL":"Total Population" ,
          "SP.POP.1564.TO.ZS":"16-64 age % Percentage of population",
          "LP.LPI.INFR.XQ":"Logistics performance index: Quality of trade and transport-related infrastructure (1=low to 5=high)",
          "EG.USE.COMM.FO.ZS":"Fossil fuel energy consumption (% of total)",
          "EG.FEC.RNEW.ZS":"Renewable energy consumption (% of total final energy consumption)",
          "EG.IMP.CONS.ZS":"Energy imports, net (% of energy use)",
          "EN.ATM.METH.KT.CE":"Methane emissions (kt of CO2 equivalent)",
          "EN.ATM.CO2E.KT":"CO2 emissions (kt)",
          "AG.LND.FRST.ZS":"Forest area (% of land area)",
          "EN.ATM.GHGT.KT.CE":"Total greenhouse gas emissions (kt of CO2 equivalent)",
          "NE.IMP.GNFS.ZS":"Imports of goods and services (% of GDP)",
          "NV.AGR.TOTL.ZS":"Agriculture, forestry, and fishing, value added (% of GDP)",
          "NE.EXP.GNFS.ZS":"Exports of goods and services (% of GDP)",
          "NY.GDP.PCAP.PP.CD":"GDP per capita, PPP (current international $)",
          "EN.ATM.NOXE.KT.CE":"Nitrous oxide emissions (thousand metric tons of CO2 equivalent)"

          }

# Store data in pandas. This  will download all requested idicators, for all     countries
df2 = wbdata.get_dataframe(indicators1, country='all', convert_date=True)

Answer 1

country_info  = wbdata.get_country(display=False)
data = {}
for i in range(len(country_info)):
    country_id = country_info[i]['id']
    try:
        df = wbdata.get_dataframe(indicators1, country=country_id)
        print ("Retrieved {0} record for country {1}".format(len(df), country_id))
        data[country_id] = df
        except:
            print ("No records for country {0}".format(country_id)) 

• Get all the country codes
• Read information by country level and push it into data dictionary

You can also create a list of all the country id's you are interested in and pass it to get_dataframe in a single call.