The Microsoft Finance Time Series Forecasting Framework, affectionately dubbed Finn, is an automated solution designed for crafting precise financial forecasts. While it primarily caters to corporate finance, its versatile capabilities make it adept for any time series forecasting challenges you might face.

Why Choose the Finn Framework?

• Automated feature engineering and selection.
• Extensive model selection with access to over 25 models.
• Integration with Azure for parallel time series processing.
• Flexibility in forecasting frequency – daily, weekly, monthly, quarterly, or yearly.
• Ability to handle both historical and future values for regressors.

Installation Process

Before diving into using the framework, it needs to be installed. Here are the steps to get you started:

1. Installing the CRAN Version

For the standard version available on CRAN, run the following command in R:

install.packages("finnts")

2. Installing the Development Version

If you’re looking for the latest features or bug fixes, the development version can be installed directly from GitHub:

install.packages("devtools")

devtools::install_github("microsoft/finnts")

How to Use the Finn Framework

Once installed, it’s time to harness its capabilities. Here’s a step-by-step guide on using the framework:

1. Load the Required Library

library(finnts)

2. Prepare Your Historical Data

Data preparation is crucial. Using the `timtk` and `dplyr` packages, you can format your historical data with the following commands:

hist_data <- timetk::m4_monthly %>%
  dplyr::rename(Date = date) %>%
  dplyr::mutate(id = as.character(id))

3. Forecast Using Finn

Now you’re ready to call the main modeling function. Think of this like sending your financial ‘ship’ (data) sailing into the forecasting ocean:

• Input Data: This is like your ship’s crew, essential for a successful voyage.
• Combo Variables: Keys to navigating the waters, which may include identifiers.
• Target Variable: The destination you want to reach – the forecasted values.
• Date Type: The type of ocean currents, i.e., monthly or weekly.
• Forecast Horizon: How far ahead you wish to look, similar to how far a ship can see over the horizon.

Here’s the command to initiate the forecasting:

finn_output <- forecast_time_series(
  input_data = hist_data,
  combo_variables = c(id),
  target_variable = value,
  date_type = month,
  forecast_horizon = 3,
  back_test_scenarios = 6,
  models_to_run = c(arima, ets),
  run_global_models = FALSE,
  run_model_parallel = FALSE
)

Troubleshooting

If you encounter any issues, here are a few troubleshooting steps you can take:

• Ensure all required packages are installed and up-to-date.
• Check your historical data format; it should align with the model's expectations.
• Consult the GitHub repository for any known issues or updates.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

In Conclusion

Finn is a powerful tool for anyone engaged in financial forecasting or time series analytics. With a bit of preparation and familiarity, you can create robust forecasts that will guide your financial decisions.

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.