Function reference • fHMM

Data preparation

Use these functions to prepare or simulate financial data.

set_controls() print(<fHMM_controls>): Set and validate controls

download_data(): Download financial data from Yahoo Finance

prepare_data(): Prepare data

fHMM_parameters() print(<fHMM_parameters>): Set and check model parameters

fHMM_data() print(<fHMM_data>) summary(<fHMM_data>): Constructor of an fHMM_data object

plot(<fHMM_data>): Plot method for an object of class fHMM_data

Model estimation

Use these function for model estimation and state decoding.

fit_model() print(<fHMM_model>): Model fitting

fHMM_model(): Constructor of a model object

decode_states(): Decode the underlying hidden state sequence

reorder_states(): Reorder estimated states

Model evaluation

Use these functions to evaluate a fitted model.

coef(<fHMM_model>): Model coefficients

npar(): Number of model parameters

compare_models(): Compare multiple models

compute_residuals(): Compute (pseudo-) residuals

plot(<fHMM_model>): Plot method for an object of class fHMM_model

fHMM_colors(): Set color scheme for visualizations

fHMM_events() print(<fHMM_events>): Checking events

predict(<fHMM_model>): Prediction

residuals(<fHMM_model>): Residuals

Data

The following data sets are included in the package.

dax: Deutscher Aktienindex (DAX) index data

spx: Standard & Poor’s 500 (S&P 500) index data

unemp: Unemployment rate data USA

vw: Volkswagen AG (VW) stock data

Models

The following pre-computed models are included in the package.

dax_model_2n: DAX 2-state HMM with normal distributions

dax_model_3t: DAX 3-state HMM with t-distributions

dax_vw_model: DAX/VW hierarchical HMM with t-distributions

sim_model_2gamma: Simulated 2-state HMM with gamma distributions

sim_model_4lnorm: Simulated 4-state HMM with log-normal distributions

unemp_spx_model_3_2: Unemployment rate and S&P 500 hierarchical HMM

Small utilities

These functions may be useful beyond {fHMM}.

check_date(): Check date format

find_closest_year(): Find the closest year to a given date

is_number(): Check for integers

is_tpm(): Check for transition probability matrix

match_all(): Best-possible match of two numeric vectors

sample_tpm(): Sample transition probability matrices

simulate_markov_chain(): Simulate Markov chain