Project that works on Forecasting, VAR model, Time series, Dynamic regression, RMarkdown
Budget: $30 – $250 USD
I need this to be done by 11th April. You have to be consistent with the course topics.
Questions:
1) In a dynamic regression model, it may make sense to include lagged variables as exogenous regressors. In the model of DK1 prices, include both contemporaneous and lagged carbon permit prices. How does this change your model? (You may want to read Ch 10.6 in fpp3). (don't need this. I already completed it)
2) From ENTSOE-E (https://transparency.entsoe.eu/) or statnett (https://www.statnett.no/for-aktorer-i-kraftbransjen/tall-og-data-fra-kraftsystemet/last-ned-grunndata/), download hourly consumption data for Norway for 2017 and 2018 (files in the attachment). Join this with the 2019 data in order to create one long time series for Norwegian consumption. Then model the seasonality in the data (at monthly, weakly and daily level), with fourier terms.
3) Create a VAR model for consumption and prices in 2019 using Danish data (You can find it on ENTSOE_E or at the Danish TSO’s energy data site (https://www.energidataservice.dk/tso-electricity/elspotprices). Create a 30 day forecast. Load in actual data for january 2020–how does your forecast look? Include wind power in Denmark as a variable. How does this affect the model and forecast?
I also added the work of question 1 named assign6.1. You don't need to work on this question, but make sure to maintain consistency. So you are working on the last 2 questions only.
Here are the relevant links for the course:
1) Main page: https://jmaurit.github.io/analytics/ (Actually you don't need this but I am adding this for reference. We will work on Lab 4-6).
2) Project questions here (at the bottom): https://jmaurit.github.io/analytics/labs/lab6.html
I need this to be done by 11th April. I will pay 70 USD.
Questions:
1) In a dynamic regression model, it may make sense to include lagged variables as exogenous regressors. In the model of DK1 prices, include both contemporaneous and lagged carbon permit prices. How does this change your model? (You may want to read Ch 10.6 in fpp3). (don't need this. I already completed it)
2) From ENTSOE-E (https://transparency.entsoe.eu/) or statnett (https://www.statnett.no/for-aktorer-i-kraftbransjen/tall-og-data-fra-kraftsystemet/last-ned-grunndata/), download hourly consumption data for Norway for 2017 and 2018 (files in the attachment). Join this with the 2019 data in order to create one long time series for Norwegian consumption. Then model the seasonality in the data (at monthly, weakly and daily level), with fourier terms.
3) Create a VAR model for consumption and prices in 2019 using Danish data (You can find it on ENTSOE_E or at the Danish TSO’s energy data site (https://www.energidataservice.dk/tso-electricity/elspotprices). Create a 30 day forecast. Load in actual data for january 2020–how does your forecast look? Include wind power in Denmark as a variable. How does this affect the model and forecast?
I also added the work of question 1 named assign6.1. You don't need to work on this question, but make sure to maintain consistency. So you are working on the last 2 questions only.
Here are the relevant links for the course:
1) Main page: https://jmaurit.github.io/analytics/ (Actually you don't need this but I am adding this for reference. We will work on Lab 4-6).
2) Project questions here (at the bottom): https://jmaurit.github.io/analytics/labs/lab6.html
I need this to be done by 11th April. I will pay 70 USD.