Truth About NICS Checks

[cjwils]

When you look at the OP's graph from late 2015 through the latest date, there is clearly a downward trend. (Looking at the actual line, not the green and red band.) I think we need to see where the line goes from here in order to see if changing presidents is correlated with a change in the long term trend line.

 

[denton]

Containing an "exponent" does not make a trend exponential. It makes it a power factor, or N order polynomial.

An exponential function has the variable AS an exponent. An n order power function has the variable to the power of n. Y = A*e^X is indeed an exponential function, because the variable, X, is in the exponent. Y = X^2, a second order polynomial POWER FUNCTION, is NOT an exponential function, as the variable is NOT a component of the exponent.



Yes, indeed you can. Functions including only fixed exponents - aka constant exponents - are indeed NOT exponential functions. That's the very definition of exponential functions as compared to that of power functions.

Go back to high school.

Having thought about it, I believe you are correct on the definition of an exponential expression.

But don't get too cocky.

The explanation that you refer to as "gibberish" is anything but. It is the very soul of curve fitting. The whole idea is to see if you can reduce the error (sum of squares per degree of freedom) in your model by finding an underlying pattern rather than simply lumping data and taking the mean.

In this case, you really only have two choices: lump the data and accept more error, or apply a trend and accept less error. Your choice. Don't let me push you one way or the other.

In order for a trend to exist, and to be found, there is no requirement whatever that the raw data not contain points that are statistically different from the chosen trend. That's simply a wrong assumption on your part.

 

[hso]

So exactly why is that so?

denton,

If you want to show an effect you have to demonstrate that there's data not reflective of it. IOW, you'd want normalized data on NICS starting before and going through the most current year to see if there was any effect dependent upon a time period (a president's term). Without data on either side of the time period you can't see a change across it.

The annual cycle of NICS checks reflects what the firearms industry sees in sales during the course of the year. At Father's Day firearms sales, and background checks, will drop until fall and pick back up as part of the normal cycle. It is perturbed by events like crimes or terrorist attacks, but the ebb and flow of firearms sales is well know in the industry.

 

[CoalTrain49]

You can clearly see the tremendous anomaly caused by Obama's election, and you can see a seasonal pattern with slower sales in the summer, and higher sales in the winter and early spring.

Which election, 2008 or 2012?

It's a well known fact that Obama’s first-term record was more focused on gun rights than control. In 2010, he signed bills allowing Amtrak passengers to pack guns in their luggage and carry loaded firearms into national parks.

He didn't start his anti-gun campaign until Sandy Hook in 2012.

More anti-gun legislation is passed by state legislators than by congress or the president yet nobody ever talks about that. It's entirely possible that many people are buying more guns in a reaction to state AG laws that are being passed or concern for self defense.

Obama campaigned for more gun control in 2012 and Trump campaigned for less in 2016 yet nothing much has changed at the national level.

 

[zb338]

I go with Denton. He isn't the only one in the world that said guns sales are, in general, on the up-swing.

Zeke

 

[denton]

denton,

If you want to show an effect you have to demonstrate that there's data not reflective of it. IOW, you'd want normalized data on NICS starting before and going through the most current year to see if there was any effect dependent upon a time period (a president's term). Without data on either side of the time period you can't see a change across it.

The annual cycle of NICS checks reflects what the firearms industry sees in sales during the course of the year. At Father's Day firearms sales, and background checks, will drop until fall and pick back up as part of the normal cycle. It is perturbed by events like crimes or terrorist attacks, but the ebb and flow of firearms sales is well know in the industry.

You are absolutely correct that it is necessary to have something to compare against. But there are many somethings that are available. For example, if you like, you can compare November 2012, when there was a national election, to the ten preceding months, when there was not and ask if the election month is statistically different. (It is.) You can also compare November 2012 vs. months following, in which there was no national election. Or any of several other possible bases.

You are also correct that it is possible to refine analysis of the data be de-trending it and de-seasonalizing it. But to what purpose?

So perhaps I should have phrased my still unanswered question a little better: Why is pre-2004 data the only allowed basis for comparison?

I have no doubt that Sandy Hook was a great factor in the panic/rush. But Sandy Hook happened halfway through December, 2012 and December was a record breaking, all-time high, statistically different (from the other months) point for NICS checks. It's hard for me to believe that Sandy Hook by itself set that record in only half a month.

 

[hso]

denton,

I frequented 3 gun shops locally and set up at 2 gun shows during that window. I repeatedly heard people rushing in to buy firearms the days after the Sandy Hook tragedy specifically because of the murders. This phenomenon occurs after ever dramatic random murder in a locale according to the gunshop owners, but compound that fear of "I need a gun because of what just happened" with " I better get one (or more) before they're banned". The first is a local response to random violent crime and the latter is a politically motivated response to prohibition. Isolated threatening events like random murders and potential political prohibitions in response always make people feel threatened and buying firearms is one response.

 

[denton]

denton,

I frequented 3 gun shops locally and set up at 2 gun shows during that window. I repeatedly heard people rushing in to buy firearms the days after the Sandy Hook tragedy specifically because of the murders. This phenomenon occurs after ever dramatic random murder in a locale according to the gunshop owners, but compound that fear of "I need a gun because of what just happened" with " I better get one (or more) before they're banned". The first is a local response to random violent crime and the latter is a politically motivated response to prohibition. Isolated threatening events like random murders and potential political prohibitions in response always make people feel threatened and buying firearms is one response.

Well said.

 

[CapnMac]

politically motivated response to prohibition.

Which is actually a key inde the graph is missing. With the caveat, that certain events created a surety that political change would occur (i.e., Orlando, Sandy Hook, Aurora, etc.)

 

[Tirod]

Regardless of what some might be concerned about statistically - the fact remains that every gun made annually is sold. And that more guns have been getting made annually than the previous year.

Add all the LEO trade ins, too. Those add to the overall sales of guns going to the public.

In point of fact, there are more people now in the US than 25 years ago, and 25 years of gun making and selling. More citizens buying all the guns the cops and agents used to own, too. If there is one down turn, it's when Clinton had millions of 1911's and M14's scrapped into tiny pieces. Overall - more guns, more people, more armed citizens. Millions more.

America has more guns in the hands of citizens than any other nation on earth. Squirm about the polynomials all you want, it's a fact. It was a serious consideration and known to our enemies, even in 1940.

Behind every blade of grass.

 

[dogtown tom]

Regardless of what some might be concerned about statistically - the fact remains that every gun made annually is sold.....

Who told you that?
If "every gun made" actually sold, Remington and others wouldn't be laying off employees, mom and pop gun stores wouldn't be closing and the industry wouldn't be offering rebates.