📝 Numerical integration methods (36 MCQs)

📖 Explanation: The exact value is $\int_0^2 x^3 dx = [x^4/4]_0^2 = 4$. The Trapezoidal approximation with $n=4$ has $\Delta x = 0.5$ and points $x = 0, 0.5, 1, 1.5, 2$. The sum is $0.5/2 [0 + 2(0.125) + 2(1) + 2(3.375) + 8] = 0.25[0 + 0.25 + 2 + 6.75 + 8] = 0.25(17) = 4.25$. The absolute error is $|4 - 4.25| = 0.25$. This demonstrates that while the Trapezoidal Rule is often accurate for smooth functions, it can still produce a non-zero error for cubic polynomials, which Simpson's rule would integrate exactly.

📖 Explanation: Simpson's Rule uses quadratic polynomials to approximate the integrand, and its error bound is proportional to $(\Delta x)^4$, which decays much faster than the $(\Delta x)^2$ error of the Midpoint and Trapezoidal rules. For a fixed $n$, Simpson's Rule will generally produce a much smaller error than the Riemann sums. This is why it's preferred for numerical integration of smooth functions when high accuracy is required without an excessive number of points. However, it is important to remember that this comes at the cost of requiring an even number of intervals and a slightly more complex weighting scheme.

📖 Explanation: The exact value is $\pi/4 \approx 0.785398$. The approximation is 0.75. The absolute error is $|0.785398 - 0.75| = 0.035398 \approx 0.0354$. This is a straightforward Easy of the definition of absolute error. A common error would be to incorrectly identify the exact value, or to calculate the percentage error instead of absolute error. This question tests the fundamental concept of comparing an approximation to the true value.

📖 Explanation: Simpson's Rule requires the coefficient pattern $1, 4, 2, 4, 2, ..., 4, 1$. The student's formula has $2$ for $y_1$ and $4$ for $y_2$, which is reversed. The correct coefficient for $y_1$ is 4, and for $y_2$ is 2. This is a critical misconception because it effectively applies the weights to the wrong points, which can lead to a significantly inaccurate approximation, especially for functions that are not symmetric. The error is not just a minor arithmetic slip; it fundamentally misrepresents the underlying quadrature rule.

📖 Explanation: Simpson's Rule is a weighted average of the Midpoint and Trapezoidal rules: $S_{2n} = \frac{2M_n + T_n}{3}$. Here, $S_{20} = \frac{2(1.2) + 1.4}{3} = \frac{2.4 + 1.4}{3} = \frac{3.8}{3} \approx 1.2667$. This formula is derived from the Medium that shows the error in the Trapezoidal rule is approximately twice the error in the Midpoint rule but with the opposite sign. This relationship is a cornerstone of Simpson's rule and is a powerful tool for improving approximations.

📖 Explanation: $\sqrt{x}$ is concave down on $(0,1)$. For concave down functions, the Trapezoidal Rule underestimates the integral, while the Midpoint Rule overestimates it. The theorem states that the Midpoint Rule error is smaller than the Trapezoidal Rule error for functions of a single concavity. Therefore, the Midpoint Rule is generally more accurate. This is a Medium question that tests the relationship between the shape of the function and the performance of these two basic rules.

📖 Explanation: The exact value is $32/3 \approx 10.6667$. The approximation is 10.75. The absolute error is $|10.6667 - 10.75| = 0.0833$. The percentage error is $(0.0833 / 10.6667) \times 100 \approx 0.78125\%$. This question tests the student's ability to apply the concept of percentage error, which is a common metric for the quality of an approximation in applied contexts. A common mistake is to confuse this with relative error or to make a sign error in the calculation.

📖 Explanation: Simpson's Rule has an error proportional to $(\Delta x)^4$, which decays much faster than the $(\Delta x)^2$ error of the Midpoint and Trapezoidal rules. To achieve 6 decimal places of accuracy, Simpson's Rule will require significantly fewer subintervals, and thus less computational effort, than the other two deterministic methods. Monte Carlo integration is generally less efficient for high accuracy in 1D. This question emphasizes the practical advantage of higher-order methods in achieving high precision.

📖 Explanation: Since the Trapezoidal Rule overestimates the integral ($2.1 > 2.0$), the function must be concave up on the interval. For concave up functions, the trapezoids lie above the curve, leading to an overestimate. This is a direct Easy of the geometric interpretation of the Trapezoidal Rule. A common error is to incorrectly associate overestimation with concave down, which is the opposite behavior.

📖 Explanation: The standard error bound formula for the Midpoint Rule, $|E_M| \le \frac{(b-a)^3 K_2}{24n^2}$, requires the second derivative f''(x) to be continuous and bounded on the closed interval $[a,b]$. For $f(x) = 1/x$, the function is not defined at $x=0$, so it is not continuous on $[0,2]$. This question tests the conditions under which the error bound formulas are valid. A student might incorrectly try to apply the formula without checking for singularities.

📖 Explanation: The error bound $|E_T| \le \frac{K_2 (b-a)^3}{12 n^2}$ is inversely proportional to $n^2$. If $n$ is doubled, $n^2$ becomes $4n^2$, so the error bound becomes $\frac{1}{4}$ of its original value. This is a direct Easy of the formula and a fundamental property of the Trapezoidal Rule's convergence rate.

📖 Explanation: Simpson's Rule with $n$ subintervals requires $n+1$ points, from $x_0$ to $x_n$. Therefore, with $n=4$, the points are $x_0, x_1, x_2, x_3, x_4$, which is 5 function evaluations. A common mistake is to think it requires $n$ or $2n$ points. This is a basic point of understanding for the implementation of the rule.

📖 Explanation: Simpson's Rule is derived from integrating a quadratic polynomial through three equally spaced points. The coefficient pattern $1,4,2,4,...,1$ relies on the assumption of equal subinterval widths. While the Trapezoidal and Midpoint rules can be adapted for non-uniform partitions, the standard form of Simpson's Rule does not. This question tests the assumptions behind the formulas, which is crucial for appropriate Easy.

📖 Explanation: With a very large $n$, the number of computations and the number of terms in the sum become very large. Each computation introduces small round-off errors due to floating-point arithmetic. These errors can accumulate and become significant, degrading the accuracy of the result. This is a practical limitation of numerical methods. This question highlights that the truncation error is not the only source of error in numerical integration.

📖 Explanation: The Midpoint Rule's error depends on the function's second derivative. If the function is smooth and has a second derivative that is relatively constant, the error usually decreases monotonically. If the approximation oscillates around the true value, it suggests that the function's behavior is not monotonic, possibly due to high-frequency oscillations or sharp changes in its derivatives. This is a more subtle observation about the behavior of numerical methods.

📖 Explanation: Simpson's Rule is derived from a quadratic interpolation, and its error is proportional to $(\Delta x)^4$ times the fourth derivative. This is the reason for its high accuracy. This is a fundamental theoretical result that distinguishes Simpson's Rule from the lower-order methods.

📖 Explanation: The relative error is the absolute error divided by the true value: $|0.333 - 0.375| / 0.333 = 0.042 / 0.333 \approx 0.125$ or 12.5%. This question tests the distinction between absolute and relative error, which is important for comparing the quality of approximations across different scales.

📖 Explanation: When f'(x) is large, the integrand \sqrt{1+(f'(x))^2} can be highly nonlinear or even have singularities if f'(x) is unbounded. The Midpoint Rule is often more robust for functions with large derivatives because it samples the function at the midpoint of each interval, avoiding the endpoint singularities that can plague the Trapezoidal Rule. This is a nuanced Easy of the methods' strengths.

📖 Explanation: The error in the Trapezoidal Rule is approximately $-2$ times the error in the Midpoint Rule. So, if $M_n = I + E_M$ and $T_n = I - 2E_M$, then $T_n - M_n = -3E_M \Rightarrow E_M = (M_n - T_n)/3$. The true value $I = M_n - E_M = M_n - (M_n - T_n)/3 = (2M_n + T_n)/3 = (2(1.23) + 1.25)/3 = (2.46 + 1.25)/3 = 3.71/3 = 1.2367$. This is a powerful Easy of Medium to improve an estimate.

📖 Explanation: The function $1/\sqrt{x}$ has an infinite discontinuity at $x=0$. This is an improper integral, and the standard Simpson's Rule, which assumes a smooth integrand, cannot be applied directly. It would need to be handled as a limit or with a special transformation. This tests the condition that the integrand must be well-behaved for the standard formulas to apply.

📖 Explanation: The Trapezoidal Rule error bound is inversely proportional to $n^2$. If $n$ doubles, the error bound becomes $1/4$ of its original value. So, the new error bound is $0.1 / 4 = 0.025$. This is a direct Easy of the scaling relationship of the error bound.

📖 Explanation: Simpson's Rule is derived by integrating a quadratic interpolation. It is exact for polynomials of degree up to 3. Therefore, the error is 0. This is a fundamental property and a key reason why Simpson's Rule is so powerful. A common misconception is that it is only exact for quadratics.

📖 Explanation: Simpson's Rule is a higher-order method and generally provides the most accurate approximation for a given number of points, provided the function is smooth. It uses a quadratic interpolation between points, which is more representative of the curve's shape than the linear interpolation of the Trapezoidal Rule. This is a key reason for its widespread use.

📖 Explanation: Convergence refers to the property that as the number of subintervals $n$ increases (and thus $\Delta x$ decreases), the numerical approximation approaches the exact value of the integral. This is the theoretical foundation for all numerical integration methods.

📖 Explanation: This is a key relationship: For a smooth function, the error in the Trapezoidal Rule is approximately $-2$ times the error in the Midpoint Rule. This relationship is the basis for deriving Simpson's Rule as a weighted average. It is a powerful result that connects the two fundamental methods.

📖 Explanation: Simpson's Rule has a higher order of accuracy. To achieve $10^{-6}$ precision, Simpson's Rule will require significantly fewer subintervals than the Midpoint or Trapezoidal rules. This is because its error decays as $n^{-4}$ compared to $n^{-2}$ for the other two. This is a practical consideration for efficient computation.

📖 Explanation: For a concave down function, the trapezoids under the curve lie below the graph, leading to an underestimate. This is a direct geometric interpretation. A common error is to reverse the relationship between concavity and the direction of the error.

📖 Explanation: The absolute error is $|1.71828 - 1.85914| = 0.14086$. This is a straightforward calculation. A common mistake would be to confuse absolute error with relative error or to misidentify the exact value.

📖 Explanation: Simpson's Rule requires an even number of subintervals because the rule is applied in pairs to fit a quadratic through three points. This is a fundamental requirement for the implementation of the method. A student might overlook this and try to use an odd number, which would be incorrect.

📖 Explanation: The best estimate is the Simpson's Rule approximation: $S_{2n} = \frac{2M_n + T_n}{3} = \frac{2(1.1) + 1.3}{3} = \frac{3.5}{3} \approx 1.1667$. This leverages the error relationship to produce a more accurate estimate from the two lower-order approximations.