There were only two things I did in Troy that could not really be replaced 100% remote. One was community meetings, the other was periodic ride alongs. These are nice to have things but are non-essential to the job. (And periodic on-site can probably meet that need.)

It ultimately takes strong leadership to accept the input of information and change behavior (as well as competent analysts on the other end). It certainly is not impossible though. Many rural agencies already have this model, with an external center providing intel and analytics.

I don’t view the police as much different than other agencies I have been a part of. On occasion I meet unreceptive people at my current job too. Sometimes I can turn resistant people over (and that resistance can be due to not wanting change, or having a low opinion of analytics), and sometimes I need to move onto other things. But most people just have general problems and are receptive to any help they can get.

]]>Thinking back to when you were an analyst, working for a PD – do you think that the remote working model still applies?

During the pandemic I found that I could do the “work” part of the equation quite well in a WFH setting; but information sharing and maintaining legitimacy with the troops was more difficult, especially working with a police agency with high turnover / extremely young officers and first-level supervisors.

Police agencies are still a hard nut to crack when it comes to reconciling the analysis/intelligence environment with reality. Realizing that we have a more agile, less likely to conform to the 40 hours of “butts in seats” workforce, we’ll need to compromise to locate and retain talent.

]]>Outcome^k_t ~ f(t,other_time_varying) # The trajectory equation for group k

Prob(K) ~ f(time_static)

So you have two equations — one estimating the ultimate outcome on the first line (the trajectory), and one estimating the probability of belonging to a particular group (the second line).

If you have constant characteristic, it will not impact the shape of the trajectory over time. It would be confounded with the intercept term in the trajectory equation, since it is a constant over time.

But, it could impact what group you are assigned to (e.g. males more likely to be in one group vs females). Time static covariates are specified via the “risk(…)” option, which is like a multinomial model that predicts the probability of group assignment.

]]>Oh great thank you so much!

And the tcov option is for time-varying covariate, so it cannot be use with categorical variables such as SES and sex to predict trajectories?

It is contained in the “_traj_Prob*” fields, so if you have 3 groups, you will have 3 columns post traj that are _traj_Prob1, _traj_Prob2, _traj_Prob3, etc. Those are the posterior probabilities for belonging to that group.

]]>the _traj_Group gives for each individual their class membership. Do you know if there is any way on Stata to extract information about each individual posterior probability of being in a given class instead? Is there a command that allows that ? Thank you again!

Frederic ]]>

Good point! I thought about it but this procedure drops observations with missing risk factor data, so then you don’t have the same trajectories once you add “risk(..). in the script. So in this case, it’s better to conduct regressions separately.

]]>traj, model(cnorm) var(CD*) indep(Age*) max(12) order(1 1 1 1) risk(tomtote5) …

]]>For example, to perform a multinomial regression, this is my script (predicting trajectories group membership from theory of mind, with the first group as the reference group) :

mlogit _traj_Group tomtote5, base(1)

_traj_Group is the variable you get automatically in your dataset once you perform GBTM. Its the class membership of each individuals. But there is a risk of uncertainty of assignment, so its more appropriate to take posterior probabilities of each individual for the regression. However, you dont get it automatically in your dataset..

Thank you!

Frederic Theriault

I am not sure what you mean, can you give a more explicit example regression equation?

]]>