statistics

You are currently browsing articles tagged statistics.

As we approach the end of 2013, I’m going to do my usual series of blog posts looking back at what actually happened on this little planet. This is the first, being a list of the most-read posts on this website.

There hasn’t been a lot to choose from in the last couple of years, because most of my writing is done elsewhere these days. That means some rather mundane pieces of writing, such as Weekly Wrap posts, end up on the list. That’s possibly an argument for abandoning this little exercise.

  1. Catchup posts within 36 hours was the most popular post of all, which makes no sense whatsoever because it’s routine administrivia. I suspect the visitor count has been artificially inflated somehow, though supposedly the traffic generated by spambots has already been removed.
  2. My tweets from TechEd Australia 2012′s keynote sessions, a post that was linked to from news stories that reported me having been banned from attending Microsoft’s TechEd conference. My own blog post on this issue is coming up at number 5.
  3. Guardian Australia not the droid you’re looking for, being my reaction against all the excitement generated in January 2013 by the announcement that there would soon be an Australian edition of this news masthead.
  4. My fish are dead: the black dog ate them (an explanation?), being my rather idiosyncratic announcement and discussion of the fact that I’d been dealing with a severe depression episode, published in July.
  5. Microsoft has banned me from covering TechEd, which is self-explanatory.
  6. Choosing my next media directions: you’re doing it, OK?, from May.
  7. Vodafone Australia’s new 4G network ain’t bad, being the write-up of my trial of the network which led to that conclusion.
  8. Weekly Wrap 152: LulzSec, Optus, radio and thinking stuff, which I suspect is only in the Top 10 because it mentions LulzSec.
  9. Weekly Wrap 155: Chemtrails, elitism and much thinking, ditto, chemtrails.
  10. Sydney Harbour “giant gambling den” bullshit reportage, from January.

Read the rest of this entry »

Composite image of ZDNet column headline and McAfee report title: click for ZDNet columnAs brokers of reliable information about the scale of online crime and espionage, most information security vendors would make great used car salesmen — but McAfee’s latest research finally seems to be taking the right path.

In my column at ZDNet Australia this week, I give McAfee some praise for the most recent research they’ve funded, a preliminary report from the Washington-based Center for Strategic and International Studies titled The Economic Impact of Cybercrime and Cyber Espionage that dismantles the daft idea that cyberstuff costs the global economy a trillion dollars a year.

McAfee now admits that you can’t run a small-N survey in a couple dozen large, wealthy nations — often a self-selected sample of known crime victims at that — and extrapolate the data globally.

Their new figure is “probably measured in the hundreds of billions of dollars”, although they never quite commit to one specific number…

“In the context of a $70 trillion global economy, these losses are small, but that does not mean it is not in the national interest to try to reduce the loss, and the theft of sensitive military technology creates damage whose full cost is not easily quantifiable in monetary terms,” McAfee writes.

True, but as McAfee themselves point out, this supposed cybercrime explosion is really down at the level of shoplifting. Retailers generally budget between 0.5% and 2% for pilferage and other such “shrinkage”.

I also mention my previous critical comments about various infosec vendors’ dodgy statistics — but I don’t link to them, because they were mostly published at non-CBS mastheads. So here’s a selection of stories I’ve written on this subject over the last couple of years.

Read the rest of this entry »

The 9pm EdictThe media totally misrepresents the real risks to our lives. Senator Conroy totally misrepresents the facts. Again. And the government literally makes my life more painful.

Here is episode 10 of The 9pm Edict.

You can listen to this episode below. But if you want them all, subscribe to the podcast feed, or even subscribe automatically in iTunes.

Play

For more information about the topics covered in this episode, check out Australia’s Standard for the Uniform Scheduling of Drugs and Poisons, the National Drugs and Poisons Schedule Committee’s (NDPSC) decision on codeine, especially the NDPSC record of reasons, 56th meeting 16-17 June 2009, the ABC News story on same, and some background on codeine; the Marketplace story Big companies are hoarding cash; ABC News stories about the Pope’s Easter service being clouded by the sex abuse scandal and Jews upset that the Vatican compares criticism of the Pope to the Holocaust; and Australian Bureau of Statistics publication 3303.0 — Causes of Death, Australia.

If you’d like to comment on this episode, please add your comment below, or Skype to stilgherrian or phone Sydney +61 2 8011 3733.

[Credits: The 9pm Edict theme by mansardian, Edict fanfare by neonaeon, all from The Freesound Project, as were the other sound effects. Photograph of Stilgherrian taken 29 March 2009 by misswired, used by permission.]

The 9pm Edict

A tsunami devastates Australia’s Twitter tragics. People continue to die in politically inconvenient accidents. And Dennis Shanahan is a disingenuous… you’ll find out the word I use. That’s not news, that’s just an observation.

Despite the lag, here is episode 3 of The 9pm Edict. Finally.

You can listen to this episode below. But if you want them all, subscribe to the podcast feed, or even subscribe automatically in iTunes.

Play

If you’d like to comment on this episode, please add your comment below, or Skype to stilgherrian or phone Sydney +61 2 8011 3733.

There will be a special extra episode on Friday 5 March to make up for the one we missed on Monday.

[Credits: The 9pm Edict theme by mansardian, Edict fanfare by neonaeon, all from The Freesound Project. Photograph of Stilgherrian taken 29 March 2009 by misswired, used by permission.]

OK, so I was wrong. Very wrong. And Citi was right, perhaps even conservative with their estimates of Google’s revenue turnaround.

The other day I was sceptical that Citi’s analysis showed Google suddenly reversing three years of declining revenue growth. They predicted that the revenue growth for 2009 Q4 would be 16%.

I scoffed. I figured the fat red line (right) was a more appropriate projection of the data in the graph.

Well, Google’s actual Q4 revenue growth was 17%. That’s the green dot.

Bastards.

Now Bob Bain, who actually knows how to read these numbers, has found a more detailed review. He points out that while net profit was allegedly up fivefold, Google did write off a billion dollars as “impairment of equity investments” last year. I don’t know what that means, but it sure sounds like it’d make last year look a bit wobbly.

Now, check this chart of Google’s share price for the last two years.

Broadly speaking, it shows the share price dropping as the global financial crisis kicked in, and then recovering. Apart from that tiny little downturn in the last week or two, Google’s share price is back up to where it was two years ago.

Lessons?

  1. The stock market is a long term investment, unless you really want to immerse yourself in the crazy world of the day traders.
  2. When it comes to the stock market, I haven’t the faintest fucking idea what I’m talking about, and you should ignore me.

Any questions?

It’s no secret that the high-growth venture-capital-driven Internet industry is a hype-laden circus of misinformation. But today’s Chart of the Day from Silicon Alley Insider (right) really takes the biscuit.

Here we see that since Q3 of 2006, Google’s quarterly revenue growth has steadily declined from a cancerous 70% year-on-year to under 10% for the last three quarters.

Now there’s nothing wrong with an 8% growth rate. That’s pretty much up there with China, for instance. And the only countries growing faster than China are starting from a very low base. Think of such economic powerhouses as Angola, Bhutan, Rwanda and Timor-Leste.

OK, I know countries and companies aren’t the same thing, but that’s not my point.

Look at the bottom right of that first graph.

Citi reckons that Google’s growth in the next two quarters will increase in a more-or-less straight line that’s even steeper than that most recent quarter-on-quarter increase. Based on fucking what, Citi? That’s the only quarter showing growth in the entire chart!

Surely this second graph is a more realistic projection based on the data available.

There’s precisely zero science here. I just grabbed the straight-line tool in my graphics program and drew in a red line by eye. I certainly couldn’t be arsed doing this properly with, you know, mathematics. But…

Honestly, which of these “projections” looks more realistic to you?

As I say, there’s no science here. Perhaps we can add some analysis.

What factors might be affecting Google’s potential for increased revenue growth?

On the plus side:

  • Google is operating in a marketplace which is still growing, and perhaps even growing faster than last year as we emerge from the global financial crisis. Then again, so is everyone else.
  • Google’s got a new product line coming out, those Android phone thingamabobs, and people are wetting themselves to get one. Well, at least until Apple reveals their new toy next week.
  • Google pwns the Internet.
  • What else?

But working against that:

  • Google is subject to increased competition, especially from Microsoft’s Bing and related products and services.
  • Rupert Murdoch is up to something, although admittedly he might not have the faintest idea what he’s doing.
  • Every market eventually reaches a plateau. The key bullshit image of the first dot.com boom was the classic revenue projection of a start-up: exponential growth continuing forever. Well, here in rich white Western countries, all the folks who are using the Internet are pretty much there already. There might be more volume of data, but the amount of attention span you can sell to advertisers ain’t going to rise so much.
  • What else?

We also need to remember that these are revenue figures, not profit. And that’s my core point. Simplistic “Oooh, aah!” gawking over one numerical measure without context is wankery. Why do we let analysts get away with it?

All that said, I’m no financial analyst. What would I know? So if you know better, set me straight. There’s a “Reply” box below…

Oh, and Google’s Q4 revenue figures will be announced tomorrow. So we’ll have one more real data point then. Stand by.

Yesterday I wrote an article for Crikey plus a post here based on Google Trends data which, it now appears, is dodgy.

Google Trends shows a steady decline in traffic to various websites since about September 2008, based on the metric “unique daily browsers”. But I was howled down. Everyone else’s metrics were not showing such a decline.

Indeed many, such as this chart of Nielsen NetRatings’ unique dailies, provided by Andrew Hunter (@Huntzie), Head of News, Sport and Finance at ninemsn, showed the exact opposite.

Nielsen NetRatings unique daily browser chart, showing steady rise in audiences: click to embiggen

For example, news.com.au grew from 250,829 average daily unique browsers (UBs) in July 2008 to 346,367 in October 2009, a 38% increase. Not the roughly 50% drop shown by Google Trends.

Google says:

Trends for Websites combines information from a variety of sources, such as aggregated Google search data, aggregated opt-in anonymous Google Analytics data, opt-in consumer panel data, and other third-party market research. The data is aggregated over millions of users, powered by computer algorithms…

In other words, it’s some Google Secret Sauce. But has the sauce gone off?

Photograph of an angry goose

The Google Trends forum is rather quiet. There were only three questions or comments posted for the whole of September, none of which received a reply, and nothing since. I can’t see that anyone from Google has responded to anything for months and months — I gave up looking back any further. Others have noted that Google Trends data differs wildly from Google’s own Analytics product — usually complaining that it shows significantly less traffic.

Google Trends is a Google Labs product, i.e. an experiment, I’m starting to think that it’s been abandoned and we’re just seeing a slow degradation due to lack of maintenance.

Meanwhile, I have changed my Twitter avatar to a goose for the rest of today.

Google Trends graph showing traffic drop to major Australian news sites

If Google Trends’ statistics are to be trusted, it looks like there’s been a significant decline in traffic to websites over the last year — not just news, but everywhere. Except social network sites.

Following a blog post by Nicholas Moerman, a planning intern with Proximity in London, I checked out the figures for Australia sites. It does indeed look like there’s been a significant drop in daily unique visitors — which is what Google Trends measures, rather than the more common monthly uniques.

Crikey logo

I’ve written more, and provided more graphs, in a piece for Crikey today, Is social media killing the web as we know it?

Oh, and I was also in Crikey yesterday, Baffled by Murdersoft? Making sense of Murdoch and Microsoft, where I look at some of the numbers behind the rumoured deal between News Corporation and Microsoft’s Bing search engine.

Stilgherrian’s links for 08 November 2009 through 18 November 2009:

See what happens when you don’t curate your links for ten days, during which time there’s a conference which generates a bazillion things to link to? Sigh.

This is such a huge batch of links that I’ll start them over the fold. They’re not all about Media140 Sydney, trust me.

Read the rest of this entry »

Stilgherrian’s links for 22 October 2009 through 27 October 2009, published after far too long a break. I really, really do need to work out a better way of doing this…

« Older entries