Package: tsn 1.4.1

Introduction | Data | Transition Network Analysis | Transition probabilities with ts_tna() | Transition frequencies with ts_ftna() | State co-occurrence with ts_cna() | Attention-weighted transitions with ts_atna() | Indexing the model with series_networks() | Local direction with trend() | Visibility with vg() | Window similarity with tsn()

Introduction | Data | Discrete-state plots | Overlay | Ribbon | Heatmap | Stack | Local-direction plots | Points | Panels | Constructed-network plots | Network | Source series | Weighted distance network | Adjusting the network encoding | Transition Network Analysis plots | Combined | Extracted-model plots | Selecting a plot

Introduction | From measurements to a model | The bridge: this object is a Nestimate model | Four constructors, one interface | One series, and what it can support | Choosing states that mean something | One network per person | Describing the model | Centrality | How much of the sequence is structure, and how much is noise | Pruning | Bootstrap: which edges survive resampling? | Stability: would the same states rank first in a smaller study? | Markov order: is a first-order model enough? | Comparing two periods | More than two groups: one model per condition | Plotting a model | Where the line falls

The question | One call, one network per condition | Why not subset the data and build each model by hand? | First failure: the subset welds broken time back together | Second failure: each subset learns its own thresholds | The collection is data | Composition and dynamics | Composition | Mean first passage times | Entropy and centrality | Drawing the networks | Are the differences real? | The permutation test | The metric panel | Bootstrap and Bayesian certainty per group | Is first order enough, per group? | Stability under case dropping | Pruning the collection | What to watch for | When to use which | References

One object, two packages | Reading the model as data | Where the process spends its time | Which states and edges carry the structure? | How the process moves | Mean first passage times | Persistence, return, and sojourn | Is the chain well-behaved? | How predictable is the process? | Does history beyond one step matter? | Simplifying the model | How sure are we about the estimates? | Sequence bootstrap | Bayesian edge certainty | Whole-network statistics: the vertex bootstrap | Case dropping, in two flavours | Split-half reliability | Comparing two models | The difference itself | Four more verdicts on the same pair | Link prediction | Grouped models | What does not apply, and why | The compatibility map | References