You're viewing all posts tagged with statistics

For those who are still stumped by the Math HL Prelim 2010 P2 Q3 involving P( |X-2| > a ) = 0.25 it’s somewhat related to Math HL Prelim 2008 P3 Q5iii that I mentioned a few months back about the use of Solver, and can be handled similarly without having to think (ahem) about the bounds being x standard deviations away from the mean and hence assume that it’s negligible.

I guess this specific concept is actually in HL Core Statistics; the earlier parts of the 2008 question were decidedly not so that entire question was placed in P3.

I was just checking my Analytics reports and am quite amused to find that I’m still getting traffic from search engines for the keyphrase “Alistair Chew”. 

Feature: SMRT Corporation’s numbers

zm the studious bloke: (via One Less Car back in January 2008)

from the annual report, passenger trips have increased 11% from 2002 – 389.7m to 434.9m. However, car kilometres operated dropped from 81.4m to 77.1m. SMRTC has been aggressively cutting costs by cramming as much as people into each train and reducing frequencies since 2004, to maximise shareholder value.

Having been intrigued by this tidbit of information, I just spent probably two to three hours poring through the past 10 years of SMRT Corp’s annual reports. There are some interesting things. 

Firstly, here are two charts with two very important things: ridership (in millions), car kilometres operated (in millions), and EBITDA per passenger carried and per car km operated, which is only available up till FY2007 – if there’s an EBIT amount for Railway operations only after that, please let me know, I might have missed it because my brain is fried. 

Clickthrough for full size, this one is a little squished. NB Railway EBITDA means earnings before interest, tax, depreciation, and amortisation of the MRT System excluding interest and investment income. 

We immediately notice a huge growth in car kilometres operated in FY2002 and FY2003.

FY2002 has the Changi Airport extension opening - remember though that back then it was a through service, it was only converted to a shuttle service to Tanah Merah in FY2004. Dover station also opened! However, this was probably at the worst of the Asian financial crisis, and ridership is said to have increased very slightly in 1H02 and actually declined in 2H02, but does not decline overall for the financial year. 

In FY2003, there’s an even larger jump of car km operated of about 10% which isn’t really explained. What we do know is that SARS started popping up in the region in 2H03, but it really only hits Singapore at the start of FY2004. I’m not sure what the impact of converting service on the Changi Airport extension to a shuttle is on car km operated – it definitely should have had a positive effect on service frequency for the stations eastward of Tanah Merah. 

We also notice that the Railway EBITDA per car km operated drops from FY2001 to FY2003, just as the car km operated goes up. Fair enough, and that’s what One Less Car was pointing out. 

What I would like to draw your attention to is FY2010. It’s taken 7 years, the Boon Lay extension opening in Feb 2009, and the Circle Line Stage 3 (5 stations, 5.7km) opening in May 2009 to surpass the same car km operated level as FY2003. No wonder SMRT can keep announcing x more trips per week, because they have the capacity – a confluence of labour and capital equipment and all the other things. 

Or so we might easily think. Trains have been out for refurbishment though - the first-generation Kawasaki C151 trains that have been in service since 1987 had their refurbishment programme ongoing from FY2006 (June 2005) to FY 2009. That may have suppressed the capacity of the system, depending on how they scheduled the refurbishment. There are also ongoing overhauls of those trains, the sixth of which was completed around the end of FY2009. 

The second-generation Siemens C651 trains (introduced 1994) are due to refurbishment by 2011 and the third-generation Kawasaki C751B (introduced 2000) will probably be up for refurbishment later in the decade, which will also have an impact on the system capacity on the North-South and East-West lines. But they’re getting new trains right? It’s supposed to be a 15% increase in capacity on the NS/EW lines with the 22 new trains and the modified Jurong East station. 

The NEL is also slated to get a 50% (seriously?!) increase in capacity eventually. Just as well, because in the past 6 years I’ve seen the NEL really fill up. Used to be able to get a seat most of the way from Outram Park to Punggol, but nope, the past few years I’ve barely been able to get a seat 3 or 4 stops away from Punggol, and that’s not during the evening peak! 

In any case, the system definitely needs these and signalling system improvements that will probably take the better part of a decade to complete, to bring the headways down some more. Let’s not even wonder whether designing the Circle Line as a medium-capacity line is a problem or not until it finally finishes opening. The next decade should be interesting as the other lines finally start coming up.. hopefully on schedule. 

Finally, here’s an insightful comment at YB on a post about the nation’s urban planning, two months ago: (emphasis is mine)

To give another analogy, metropolitan Paris within the peripherique is equivalent in area to Raffles Place extended out to Toa Payoh. (or not much further beyond) and it is humanly possible to walk through most of Paris in an entire day. My point is that as a result of decisions and choices made in the early days of Singapore’s physical development, the scale of Singapore’s planning and where most Singaporeans live, almost everyone who cannot afford a car needs to rely on public transportation (over long distances) to get to where they need to get to.

The infrastructure for pedestrians is purely utilitarian – in terms of covered walkways, air-conditioned underpasses in the shopping district – is purely a means of accessing the main stations. As such, walking as a means of transport is not enabled. (Because we do not have tight streets and blocks lined with shops, sidewalk interactions, etc…that used to characterise our now sanitised five-foot-ways.) By contrast, for example, New Yorkers walk a lot… and many walk long distances. (arguing about our climate is a different matter)

So in a sense, Singapore’s scale is such that it relies so much on either the expressway or the MRT (or the bus through very circuitous routes) to get from one place to another. When its population faces a sudden spurt in away that was not planned for, it becomes obvious that the entire transportation system comes under duress.

The Mathematics HL Statistics Option and your TI-84+ GDC: why you should know how to use the Solver. 

The fun part is that you actually can have multiple variables within the Solver expression, which will all appear on the next page after you press down. Slot in (n-1) values, go to the last one, and Solve it. It’s useful for all kinds of things, but I’m too tired now to cook up other examples. 

Correction (6/10/2010): the final line in solver should be 0 = normalcdf(-A,A,46.63,2.133) - 0.4.