Hacker News Reader

Comment by gravypod | Hacker News Reader

That looks more like noise than something I'd be willing to bet on. Previous years, within that same graph exhibit similar trends to what the highlighted "downward" portion shows.

View on HN

I agree. I hate that I have to preface this sort of thing by pointing out that I am not against legalization. But, according to that graph, the rate is actually higher after the legalization of marijuana, but the "trend" is down. Except that the trend would have been upwards had you made the stopped measuring at the end of 2014 instead of the end of 2015. And they're cheating by not measuring the trend in the same way for previous years as the post-marijuana years because before the graph the trend is measured as a single line spanning 14 years, while the trend is only for 2 years afterwards. So you'd need to do some sort of running average, or whatever the equivalent is for linear regression. Actually now I'm seriously interested in knowing if there's a running version of linear regression like there is for averages.

I believe you can use a polynomial curve fitter for this sort of "regression" [0]. There are other, much better, curve fitting algorithms out there than in NumPy. Some implementations can also guess how many terms there are and iteratively optimize making it very similar to a linear regression.

A good first order (that is much more commonly used) is to just, as you have already pointed out, use centered averages spanning N days of data.

edit: I'd also like to note that I'm neither pro nor anti legalization. I am definitely against false reporting, knee jerking, and lying to the public with statistics.

[0] - https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/...

The paper reports:

> Colorado’s legalization of recreational cannabis sales and use resulted in a 0.7 deaths per month (b = −0.68; 95% confidence interval = −1.34, −0.03) reduction in opioid-related deaths.

Which if I understand right, they're pretty sure it did something, but that something might be so small that the value of reporting it is negligible.

That seems to imply causation. I am skeptical.

Don't get me wrong, I smoke a lot of pot. I don't do so for health reasons. I just like smoking pot.

The default position I hold is skepticism. There have been some less than truthful claims of efficacy and curative properties.

I'm very much pro legalization. The medicinal benefits have no bearing on this opinion.

Judging by the actual data, the death count seems to have stabilized and stopped increasing at least. Unfortunately it's hard to tell with that completely bogus "trend" line atop the data.

Spurious Correlation - http://tylervigen.com/spurious-correlations

As usual for popsci reporting, the headline discards any nuance the article may have had — which itself usually goes beyond the claims made in the science. :-(

> The authors stress that their results are preliminary

And we should take this with a huge grain of salt, wait a few more years, and see what the data looks like then.

I wish it was easy for citizens to get a hold of data like this. So much good that could be done with clean access to key data sources. Will we ever see the day where this sort of information is put in the hands of people who want to make a positive difference.

Sources please?

edit

Sorry, I'm asking about the "previous years similar trends".

I believe gravypod is referring to previous years in the graph that starts the article. For instance, look at the 2008-2012 period. There is a definite downward trend.[0]

[0]: https://img.washingtonpost.com/blogs/wonkblog/files/2017/10/...

Yes, exactly. To prove a correlation, in my opinion, one would need....

    1. More smoothed/averaged dataset (centered rolling average ~4 months?)
    2. Explain the other extreme dips in this graph (2005, 2007, 2011)
    3. Plot similar data from other states (with and without legal marijuana)
    4. Investigate other causes (see sparrish's comment)

This is opinion, presumably informed by the graphic at the very top of the linked article. Asking for a source for an opinion is a little silly.

Is that a joke? The source is the linked article. The chart hardly shows a convincing downturn. The trend lines are very generously positioned.

Reply on news.ycombinator.com