What on earth do you mean by ‘the “nothing else happened”
parameter”?

Andy Freeman said:

Lambert’s model is for a transition between two stable situations with
some “noise”. He uses it to argue that gun control explains the
transition. Yet, he doesn’t bother to show whether or not anything
else happened at the relevant time, whether or not the situations were
in fact stable, and so on.

You call this a “parameter”?? This is a bizarre usage even by your
standards. Your claim that I have not shown that the situations were
stable is false. The homicide rate was roughly constant in the period
before gun control and in the period after gun control. Whatever
effects other factors may have had on the homicide rate in these
periods are lost in the noise.

I do not claim that the data proves that gun control caused the
change, but it is more plausible than any other explanation offered so
far:

  1. In the book from which I extracted the original data, Wallace
    suggests that the change was caused by a decline in infanticide.
    Indeed, there was a decline in infanticide, but not sufficient to
    explain the observed change — i.e. if we exclude infanticides from
    the data the drop is still statistically significant.

  2. It might be that demographic change caused the decline i.e. a
    decrease in the percentage of young men in the population. However,
    the demographic change associated with WWI when 40% of the males 18-45
    enlisted, is far far larger than any other change during the period,
    and the homicide rate 1915-18, while lower than the surrounding years
    is still higher than the post gun control rate. Furthermore,
    correcting for the demographic effects of WWI destroys the “gradual
    decline” model — it does not fit the adjusted homicide rates at all
    well.

  3. Perhaps changes in unemployment and poverty caused the decrease.
    This seems unlikely since the Depression does not seem to have
    affected the homicide rate at all.

  4. Maybe the decrease was caused by increasing urbanization. This
    seems unlikely since urbanization increased gradually over the period
    1900-1970, not abruptly around 1920.

  5. If some other factor is supposed to explain the decrease, it must
    explain why the rate did not decrease in the adjacent states of
    Queensland and South Australia until they introduced gun control (in
    1927 and 1929 respectively). It must also explain why the rate for
    homicide by “other means” did not decrease significantly after 1920,
    but the firearm homicide rate did.

“Ahem” indeed. The data isn’t complete, so its “fit” is basically
meaningless. For one, the step model explanation is inappropriate
without “nothing else changed” data, which hasn’t been presented.

Gee, Andy what has changed your mind? On Wed, 18 Dec 1991 Andy Freeman said:

Why doesn’t Lambert tell us about pre-control crime and murder rates
and trends in Oz and compare them to post-control rates and trends?
If gun control actually worked in Oz, the introduction of controls was
associated with a good change in the rate trend. Could it be that
there wasn’t a good change associated with the introduction of
controls?

I provide the data that Andy requested and now he tells us that it is
meaningless unless I provide data about “everything else” that
happened. If you felt that the data you asked for was meaningless,
why did you ask for it? Could it be that you consider data to be
“meaningless” only if it does not support your beliefs?