MH2500: pass the exams, not just read the notes
Your complete guide to Nanyang Technological University's probability and introduction to statistics course. See where the marks are, work real practice questions, and study with an AI tutor that knows MH2500.
Sia generates MH2500 practice questions, walks through events and bayes' theorem step by step, and quizzes you on the material the exam weights most heavily.
Worked example
A test for a condition affecting 1% of a population is 99% accurate in both directions. Someone tests positive. What is the probability they have the condition?
Set up with a concrete population, which is the fastest route through Bayes. Take 10,000 people: 100 have the condition, 9,900 do not.
Count the false positives. Of the 9,900 who do not, 1% test positive anyway, giving 99 false positives.
Divide. Positives total 198, of whom 99 actually have the condition, so the probability is 99/198 = 50%. The result is driven by base rate: because the healthy group is 99 times larger, its 1% error rate produces as many positives as the sick group's 99% success rate.
The trap: Option B is the base rate fallacy, and it is the single most consequential error in applied probability — it appears in medical screening, security alerts and fraud detection alike. The test's accuracy describes P(positive given condition); the question asks P(condition given positive). Bayes' theorem exists precisely because those two are different numbers, and NTU places it in week 1 for that reason. classic slip!
One exam decides 50% of your grade. Marked against published rubrics. This whole page is built around that.
Overview
What MH2500 is, and where it sits
MH2500 is described by NTU as a core mathematical course aiming to develop your understanding of fundamental concepts in probability and statistics such as random variables, independence, basic probability distributions, and confidence intervals. It also states the course's role in the degree: it prepares you for further statistics courses such as MH3500 in the Statistics Track.
The published thirteen-week schedule is weighted heavily toward distribution theory. Weeks 1 and 2 cover probability foundations — events, total probability, Bayes and independence. Weeks 3 to 5 are discrete random variables, weeks 6 and 7 continuous, weeks 8 and 9 jointly distributed, and weeks 10 and 11 expectations involving multiple random variables. Only the final two weeks turn to the law of large numbers, the central limit theorem and hypothesis testing.
That shape matters for planning: nine of thirteen weeks are spent on distributions before any inference appears. Students expecting a statistics course meet probability theory for most of the semester.
Always treat your own course outline and the exam timetable as authoritative.
Difficulty & time commitment
Is MH2500 hard, and how much time does it take?
MH2500 is manageable if you keep a weekly rhythm and treat the back half as the main event. The pattern is consistent: it starts gently and steepens, and the heaviest assessment is the part that separates grades.
The difficulty curve and the assessment weighting point the same way: the back half is harder and worth more. Front-loading effort there is the highest-return decision in the course.
Is this course for you
Who tends to do well, and who tends to struggle
You will likely do well if
- Your calculus is fluent; continuous and joint distributions require integration throughout.
- You apply Bayes by setting up a concrete population rather than manipulating the formula.
- You show your reasoning fully — the published rubric marks validity and clarity of argument, not just the answer.
- You participate in tutorials; 10% is small but it is the only component not sat under time pressure.
You may struggle if
- You expect a statistics course and are surprised when eleven of thirteen weeks are probability.
- You treat uncorrelated as independent, or assume independence without checking.
- You reach right answers by invalid reasoning — the rubric explicitly fails that.
- You miss a test without a Singapore-registered medical certificate, which the policy requires specifically.
- For every distribution, know its parameters, mean, variance and the situation it models. That card answers most of weeks 3 to 7.
- Practise Bayes problems with concrete counts rather than symbols; the base rate effect only becomes obvious that way.
- Write full reasoning in tests, since the rubric marks method of approach and clarity of argument separately from correctness.
- Do not defer weeks 12 and 13 — the limit theorems and inference arrive last but carry equal weight in the examination.
Syllabus
The 12 topics, topic by topic
The exam-weight marker on each topic shows where the marks concentrate. The amber topics carry the highest exam weight.
T1 · Events, probability and the law of total probability
Week 1; ILO1-2Sample spaces, probability axioms, and decomposing a probability across cases.
T2 · Bayes' theorem
Week 1; ILO6-7Reversing a conditional probability, and why it is the workhorse of applied probability.
T3 · Independence
Week 2; ILO3When events carry no information about each other, and how easily this is assumed wrongly.
T4 · Discrete random variables and distributions
Weeks 3-5; ILO3-5Mass functions and the standard discrete families, across three weeks.
T5 · Continuous random variables and distributions
Weeks 6-7; ILO3, 6-7Density functions and the continuous families, including the normal.
T6 · Jointly distributed random variables
Weeks 8-9; ILO3, 6-7Joint, marginal and conditional distributions for discrete and continuous variables.
T7 · Functions of several random variables and order statistics
Content; ILO6Deriving the distribution of a function of several variables, including order statistics.
T8 · Expectation, variance and Chebyshev's inequality
Weeks 10-11; ILO5-6Expectation as an operator, and a bound that holds for any distribution.
T9 · Covariance and conditional expectation
Weeks 10-11; ILO5-6Measuring joint variability, and expectation conditioned on information.
T10 · Moment generating functions
Content; ILO6Encoding a distribution in a transform, and reading moments back out.
T11 · Law of large numbers and the central limit theorem
Week 12; ILO8Why sample means converge, and why sums become normal.
T12 · Random sampling, estimation and hypothesis testing
Week 13; ILO9The inference material the course ends on, and the bridge to MH3500.
How it's assessed
Assessment structure
| Component | Weight | Format & timing |
|---|---|---|
| Examination (2 hours) | 50% | Two-hour examination of short answer questions, assessed against all nine ILOs. Examination period. Marked against published rubrics. |
| Test 1 | 20% | Individual test covering ILOs 1, 2, 3 and 5. Mid-semester. Medical certificate required for absence. |
| Test 2 | 20% | Individual test covering ILOs 5, 6 and 7. Later in semester. Medical certificate required for absence. |
| Tutorial Participation | 10% | Participation in tutorials, assessed against all nine ILOs on a published rubric covering participation and presentation. Weekly. Rubric-based. |
- The four components sum to 100. NTU's absence policy is specific: if you are sick and unable to attend a test you must email the instructor and submit the original medical certificate to an administrator, and the certificate must be issued in Singapore by a practitioner registered with the Singapore Medical Association.
- Ninety percent of the grade sits in timed assessment — two 20% tests plus a 50% two-hour examination — with only 10% in tutorial participation. The examination is short answer rather than multiple choice, and NTU publishes the rubric it is marked against: method of approach, validity of reasoning, and clarity of argument. Reasoning that reaches the right answer by an invalid route does not score.
This is an exam-cram course. With the exams at 90% of the grade and the examination (2 hours) alone at 50%, your result is overwhelmingly decided by how well you perform under time pressure. Marked against published rubrics.
Final exam timing: Two-hour examination during the examination period. Confirm the exact date and venue on your exam timetable.
How to actually pass it
A weekly rhythm, two checklists, and the traps to avoid
The course rewards consistency over cramming, and practice over re-reading. Here is the loop that works, then what to have nailed before each exam.
The weekly loop
Before the mid-semester checklist
- Compute probabilities using the axioms, total probability and Bayes' theorem.
- Apply independence correctly and work with discrete distributions.
- Work with continuous distributions including the normal.
- Handle joint, marginal and conditional distributions.
Before the final heaviest topics
- Derive distributions of functions of several random variables, including order statistics.
- Compute expectation, variance, covariance and conditional expectation, and apply Chebyshev's inequality.
- Use moment generating functions.
- Apply the law of large numbers and central limit theorem, and carry out estimation and hypothesis testing.
The mistakes that cost marks
Base rate ignored. P(positive given condition) and P(condition given positive) are different numbers. Bayes' theorem exists because of that gap, and it is week 1 material.
Uncorrelated read as independent. Independence implies zero covariance; zero covariance does not imply independence.
Right answer, invalid reasoning. The published rubric marks validity of reasoning separately. An answer reached by a route whose conditions are not satisfied scores at the fail standard.
Expecting statistics early. Eleven of thirteen weeks are probability. Planning around a statistics course misreads where the effort goes.
Teaching team
Who teaches MH2500
The bios below are factual. We do not rate lecturers; any star ratings are submitted by students who have taken MH2500.
Teaching team as listed in public course information. AskSia does not rate lecturers; star ratings are submitted by students who have taken MH2500.
Formula & concept sheet
The vocabulary and formulas you must own
- Law of total probability
- Decomposing a probability by summing over a partition of the sample space.
- Bayes' theorem
- The rule reversing a conditional probability using the prior and the total probability of the evidence.
- Base rate
- The underlying prevalence in a population; ignoring it is the classic error in conditional probability.
- Independence
- The condition that one event carries no information about another.
- Probability mass and density function
- The descriptions of a discrete and a continuous random variable respectively.
- Joint distribution
- The distribution of several random variables considered together.
- Marginal distribution
- The distribution of one variable obtained by summing or integrating out the others.
- Order statistics
- The distribution of sorted values from a sample, such as the minimum or maximum.
- Chebyshev's inequality
- A bound on the probability of deviating from the mean, valid for any distribution with finite variance.
- Covariance
- A measure of joint variability; zero for independent variables, but not conversely.
- Moment generating function
- A transform whose derivatives at zero yield the moments of a distribution.
- Central limit theorem
- The result that sums of independent variables are approximately normal regardless of the underlying distribution.
Common acronyms: AU · CLT · ILO · LLN · MGF.
Where it fits
Prerequisites, related courses & why it matters
Prerequisites published by NTU: MH1100 and MH1101, or MH1800 and MH1801, or MH1101 and MH110S, or MH1100 and MH111S, or MH1802, or CY1601, or MH1805. The course carries 4 Academic Units and 51 contact hours (39 lecture, 12 tutorial), offered in Year 2 Semester 1. Textbooks are Sheldon Ross, A First Course in Probability, and Ross, Introductory Statistics.
Your MH2500 study toolkit
Study the course with Sia, not just read about it
Each tool already knows MH2500: your syllabus, your texts, and where the marks are. Grouped by how you study, from first contact to exam week.
FAQ
Frequently asked questions
Is MH2500 hard?
It rates moderately hard. The material is standard but relentlessly quantitative, and 90% of the grade sits in two tests and a two-hour examination with only 10% in participation.
What is the assessment breakdown?
Examination (2 hours) 50%, Test 1 20%, Test 2 20%, tutorial participation 10%. The examination is short answer questions.
What if I miss a test?
NTU's policy is specific. You must email the instructor about the absence and submit the original medical certificate to an administrator, and the certificate must be issued in Singapore by a practitioner registered with the Singapore Medical Association.
What do I need before taking it?
One of several calculus routes: MH1100 and MH1101, MH1800 and MH1801, MH1101 and MH110S, MH1100 and MH111S, MH1802, CY1601, or MH1805.
Who coordinates the course?
The published course coordinator is Assoc Prof Wu Guohua, Division of Mathematical Sciences.
How much of it is statistics?
Less than the title suggests. The published schedule gives weeks 1 to 11 to probability and distribution theory, with random sampling, estimation and hypothesis testing arriving only in week 13.
Study MH2500 with Sia
Work through events, bayes' theorem, independence and the rest of the course with a tutor that knows it and quizzes you on the topics the assessments weight most heavily.
Start studying with Sia
