README.md

NowcastLSTM.jl

Julia wrapper for nowcast_lstm Python library. MATLAB and R wrappers also exist. Long short-term memory neural networks for economic nowcasting. More background in this paper in the Journal of Official Statistics.

Installation and set up

Installing the library

using Pkg
Pkg.add(PackageSpec(url="https://github.com/dhopp1/NowcastLSTM.jl"))
using NowcastLSTM

Install other required libraries with:

using Pkg
Pkg.add.(["CSV", "DataFrames", "PyCall"])

Python installation: The PyCall Julia library should install Python for you. To find out where this is installed, run:

using PyCall
sys = pyimport("sys")
sys.executable

Then to install required Python libraries, from the command line run:

python_path_from_sys_executable/python -m pip install dill numpy pandas pmdarima torch nowcast-lstm

to use a different Python installation:

using Pkg
ENV["PYTHON"] = "path/to/python"
Pkg.build("PyCall")

Example: NowcastLSTMjl_example.zip contains a Jupyter Notebook with a dataset and more detailed example of usage in Julia.

Background

LSTM neural networks have been used for nowcasting before, combining the strengths of artificial neural networks with a temporal aspect. However their use in nowcasting economic indicators remains limited, no doubt in part due to the difficulty of obtaining results in existing deep learning frameworks. This library seeks to streamline the process of obtaining results in the hopes of expanding the domains to which LSTM can be applied.

While neural networks are flexible and this framework may be able to get sensible results on levels, the model architecture was developed to nowcast growth rates of economic indicators. As such training inputs should ideally be stationary and seasonally adjusted.

Further explanation of the background problem can be found in this UNCTAD research paper. Further explanation and results in this UNCTAD research paper.

Quick usage

The main object and functionality of the library comes from the LSTM object. Given data = a Julia DataFrame of a date column + monthly data + a quarterly target series to run the model on, usage is as follows:

using NowcastLSTM

# note, any arguments which take Python objects or functions, like criterion, fill_ragged_edges, etc., pass like so
LSTM(..., fill_ragged_edges=pyimport("numpy").nanmean)
LSTM(..., optimizer=pyimport("torch").optim.Adam)

model = LSTM(data=data, target_variable="target_col_name", n_timesteps=12) # default parameters with 12 timestep history

model = train(model)

train_predictions = predict(model, data) # predictions on the training set

# predicting on a testset, which is the same dataframe as the training data + newer data
# this will give predictions for all dates, but only predictions after the training data ends should be considered for testing
test_predictions = predict(model, test_data) # predictions on the training set

# to gauge performance on artificial data vintages
ragged_predictions = ragged_preds(model, pub_lags, lag, test_data)

# save a trained model
save_lstm(model, "trained_model.pkl")

# load a previously trained model
trained_model = load_lstm("trained_model.pkl")

LSTM parameters

• data: DataFrame of the data to train the model on. Pass JuliaToPandas(julia_df). Should contain a target column. Any non-numeric columns will be dropped. It should be in the most frequent period of the data. E.g. if I have three monthly variables, two quarterly variables, and a quarterly series, the rows of the dataframe should be months, with the quarterly values appearing every three months (whether Q1 = Jan 1 or Mar 1 depends on the series, but generally the quarterly value should come at the end of the quarter, i.e. Mar 1), with NAs or 0s in between. The same logic applies for yearly variables.
• target_variable: a string, the name of the target column in the dataframe.
• n_timesteps: an int, corresponding to the "memory" of the network, i.e. the target value depends on the x past values of the independent variables. For example, if the data is monthly, n_timesteps=12 means that the estimated target value is based on the previous years' worth of data, 24 is the last two years', etc. This is a hyper parameter that can be evaluated.
• fill_na_func: a function used to replace missing values. Should take a column as a parameter and return a scalar, e.g. np.nanmean or np.nanmedian.
• fill_ragged_edges_func: a function used to replace missing values at the end of series. Leave blank to use the same function as fill_na_func, pass "ARMA" to use ARMA estimation using pmdarima.arima.auto_arima.
• n_models: int of the number of networks to train and predict on. Because neural networks are inherently stochastic, it can be useful to train multiple networks with the same hyper parameters and take the average of their outputs as the model's prediction, to smooth output.
• train_episodes: int of the number of training episodes/epochs. A short discussion of the topic can be found here.
• batch_size: int of the number of observations per batch. Discussed here
• decay: float of the rate of decay of the learning rate. Also discussed here. Set to 0 for no decay.
• n_hidden: int of the number of hidden states in the LSTM network. Discussed here.
• n_layers: int of the number of LSTM layers to include in the network. Also discussed here.
• dropout: float of the proportion of layers to drop in between LSTM layers. Discussed here.
• criterion: PyTorch loss function. Discussed here, list of available options in PyTorch here.
• optimizer: PyTorch optimizer. Discussed here, list of available options in PyTorch here. E.g. torch.optim.SGD.
• optimizer_parameters: dictionary. Parameters for a particular optimizer, including learning rate. Information here. For instance, to change learning rate (default 1e-2), pass Dict("lr" => 1e-2), or weight_decay for L2 regularization, pass Dict("lr" => 1e-2, "weight_decay" => 1e-3). Learning rate discussed here.

LSTM outputs

Assuming a model has been instantiated and trained with model = LSTM(...); model = train(model), the following functions are available, run ?function on any of them to find out more about them and their parameters. Other information, like training loss, is available in the trained model object, accessed via ., e.g. model.train_loss:

• predict: to generate predictions on new data
• save_lstm: to save a trained model to disk
• load_lstm: to load a saved model from disk
• ragged_preds(model, pub_lags, lag, new_data, start_date, end_date): adds artificial missing data then returns a dataframe with date, actuals, and predictions. This is especially useful as a testing mechanism, to generate datasets to see how a trained model would have performed at different synthetic vintages or periods of time in the past. pub_lags should be a list of ints (in the same order as the columns of the original data) of length n_features (i.e. excluding the target variable) dictating the normal publication lag of each of the variables. lag is an int of how many periods back we want to simulate being, interpretable as last period relative to target period. E.g. if we are nowcasting June, lag = -1 will simulate being in May, where May data is published for variables with a publication lag of 0. It will fill with missings values that wouldn't have been available yet according to the publication lag of the variable + the lag parameter. It will fill missings with the same method specified in the fill_ragged_edges_func parameter in model instantiation.
• gen_news(model, target_period, old_data, new_data): Generates news between one data release to another, adding an element of causal inference to the network. Works by holding out new data column by column, recording differences between this prediction and the prediction on full data, and registering this difference as the new data's contribution to the prediction. Contributions are then scaled to equal the actual observed difference in prediction in the aggregate between the old dataset and the new dataset.