Print

Print


*****  To join INSNA, visit http://www.insna.org  *****

Emiel,

As far as I know, you cannot put NAs in the attribute files for MPNet (Peng
or someone else from MelNet can correct me if I am wrong here). So to use
your data in MPNet you might have to do some sort of imputation for missing
data.

Unfortunately, there is no ALAAM estimation software in R (that I know of)
- specifically statnet does not have this functionality.

However I have very recently written a Python implementation of ALAAM
estimation (and simulation and goodness-of-fit), available from:

https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_stivalaa_ALAAMEE&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=yQQsvTNAnbvDXGM4nDrXAje4pr0qHX2qIOcCQtJ5k3w&m=XrwS5980_B0ppSPHcalrMQTUjBlb-DfyDd4-4FUg0v4&s=OHoz87LUU60YIHwGwy_yKnRVg7uQGQskSB983K1DsVY&e= 

This implementation allows NA values in the attribute files. Although it
was mainly written to experiment with the equilibrium expectation (EE)
algorithm (Byshkin et al. 2018) for ALAAM estimation, I also implemented
the Robbins-Monro stochastic approximation algorithm (exactly as used in
MPNet and the older IPNet program) for comparison. As the EE algorithm
implementation for ALAAM is definitely to be considered experimental at
this stage, you should probably use the stochastic approximation algorithm
for now.

Unlike MPNet, it only handles one-mode undirected networks for now, but
part of my motivation for this implementation in Python was to make it
readable as a demonstration or reference implementation, and easy to add
new change statistics etc.

This should be fine for small amounts of missing data. However, if there is
a large amount of missing data, and specifically if the data is missing
because it is a network sample rather than simply some data missing at
random, it may be more appropriate to consider it as a network sampling
problem for ALAAM (see Stivala et al. 2020)

Regards,
Alex.
Institute of Computational Science, Università della Svizzera italiana.


Byshkin, M., Stivala, A., Mira, A., Robins, G., & Lomi, A. (2018). Fast
maximum likelihood estimation via equilibrium expectation for large network
data. Scientific Reports, 8(1), 1-11.

Stivala, A. D., Gallagher, H. C., Rolls, D. A., Wang, P., & Robins, G. L.
(2020). Using Sampled Network Data With The Autologistic Actor Attribute
Model. arXiv preprint arXiv:2002.00849.



On Mon, Jun 15, 2020 at 11:22 PM Emiel de Lange <[log in to unmask]> wrote:

> *****  To join INSNA, visit https://urldefense.proofpoint.com/v2/url?u=http-3A__www.insna.org&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=yQQsvTNAnbvDXGM4nDrXAje4pr0qHX2qIOcCQtJ5k3w&m=XrwS5980_B0ppSPHcalrMQTUjBlb-DfyDd4-4FUg0v4&s=aECgtg9vg5mlBHIzktXoIJ8f5AL6fUAFrTop990yHc4&e=   *****
>
> Hi,
>
> I am trying to use MPNet to estimate ALAAMs. However, I receive an error
> message saying the input string is incorrect.
>
> I wonder if this might be because I have NAs in my attribute files. Can
> ALAAMs in MPNet handle missing attribute data?
>
> Is it possible to estimate ALAAMs in R?
>
> Many thanks,
>
> Emiel
>
> The University of Edinburgh is a charitable body, registered in Scotland,
> with registration number SC005336.
>
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (https://urldefense.proofpoint.com/v2/url?u=http-3A__www.insna.org&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=yQQsvTNAnbvDXGM4nDrXAje4pr0qHX2qIOcCQtJ5k3w&m=XrwS5980_B0ppSPHcalrMQTUjBlb-DfyDd4-4FUg0v4&s=aECgtg9vg5mlBHIzktXoIJ8f5AL6fUAFrTop990yHc4&e= ). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
>

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.