---
title: "2. Ordinary VAR"
output:
  rmarkdown::html_vignette:
    toc: true
vignette: >
  %\VignetteIndexEntry{2. Ordinary VAR}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(
  collapse = TRUE, comment = "#>", message = FALSE, warning = FALSE,
  fig.width = 7, fig.height = 5.5
)
```

```{r libraries}
library(idiographic)
data(srl)
vars <- c("efficacy", "value", "planning", "monitoring", "effort")
has_cograph <- requireNamespace("cograph", quietly = TRUE)
```

`build_var()` is the transparent OLS baseline: current variables are regressed
on an intercept and their lag-1 values. It uses the same lag preparation,
scaling, and within-person centering as `graphical_var()`, but applies no
regularization or EBIC selection — making it the natural reference point for
the regularized methods.

# Fit one person

Pass `subject` to the estimator instead of slicing the data frame yourself.

```{r var}
var_fit <- build_var(srl, vars = vars, id = "name", subject = "Grace",
                     scale = TRUE)

var_fit
```

Printing the fit *shows* both estimated networks — the directed lag-1 temporal
network and the undirected contemporaneous partial-correlation network — with
their weight ranges, so you see the model without calling another function.

# Tidy tables

Each question is one verb returning a tidy `data.frame`. Use `head()` when you
only want the strongest edges; never index the result with brackets.

```{r var-tables}
head(edges(var_fit))

nodes(var_fit)

summary(var_fit)
```

`coefs()` returns the full coefficient table (every cell, including zeros), and
`matrices()` prints the raw estimator matrices compactly:

```{r var-coefs}
head(coefs(var_fit))

matrices(var_fit)
```

# Plot

`plot()` draws the whole result; pass `layer` to draw a single network.

```{r plot-var, eval=has_cograph}
plot(var_fit)
```

```{r plot-var-temporal, eval=has_cograph}
plot(var_fit, layer = "temporal")
```

```{r plot-var-contemp, eval=has_cograph}
plot(var_fit, layer = "contemporaneous")
```