contents.qmd

---
title: "netdiffuseR at Sunbelt 2019"
author: "Thomas W. Valente and George G. Vega Yon"
date: "June 18, 2019"
output: html_document
---

# Introduction

## Today's objectives:

1.  Read data into netdiffuseR

2.  Simulate diffusion networks

3.  Use the visualization tools

4.  Use statistical inference tools


## Dynamic of work

1.  We will introduce a concept/function

2.  We will provide examples

3.  Attendees will be asked to answer a couple of questions using netdiffuseR


# Simulation of diffusion networks: rdiffnet

*   Disease spreading

*   Complex contagion

*   Network interventions: Mentor matching

*   Example simulating a thousand networks by changing threshold levels.
    The final prevalence, or hazard as a function of threshold levels.

## Problems

1.  Given the following types of networks: Small-world, Scale-free, Bernoulli,
    what set of $n$ initiators maximizes diffusion?


# Reading networks: survey_to_diffnet

*   Survey data

*   Static networks

*   From other formats

## Problems

2.  Read in all the 3 different networks objects in ... and create diffnet
    objects. A fake dynamic survey, an sns network and an igraph network.
    (hint: All 3 networks should result in the same diffnet object)

3.  With the new diffnet object, apply the same analysis as before.
    Which strategy maximizes adoption?

# Computation of exposure terms

*   Cohesive exposure

*   Structural Equivalence Exposure

*   Attribute weighted exposure

*   Attribute similarity weighted exposure

## Problems

4.  Using the network from problem 3, compute the following exposures:
    
    a.  Cohesive, and SE exposure
    b.  ABC attribute weighted
    c.  Similarity weighted exposure in gender and age.
    

# Statistical inference I

*   Moran's I

*   Permutation tests: Structural test
    
    *   Simple isomorphic
        
    *   When degree is endogenous

## Problems

5.  Using the function `moran`, replicate the moran's I statistics
    that the `summary` method reports when applied to the diffnet object
    from problem 3.
    
6.  

```{r}
library(netdiffuseR)
set.seed(1231)
net <- rgraph_ba(m=4L, t=199)
X   <- dgr(net)

altnet <- vertex_covariate_compare(net, X, "distance")
altnet@x <- 1/altnet@x
g <- rdiffnet(seed.graph = net, t=10, threshold.dist = function(x) rbeta(1, 2, 10),
              exposure.args = list(alt.graph = altnet, valued= TRUE))

```



# Statistical inference II

*   Lagged Exposure Models: OLS and others

*   Contemporaneous Exposure Models: Spatial Autoregressive Models


# Visualization of diffusion processes

*   


# Other applications



*   Bass diffusion models

*   Network bootstrapping