Chapter 24: Numerical Methods and Approximation (Set-2)

While iterating for a root, what does a “tolerance” mainly represent in practice?

A Exact root value
B Function maximum
C Interval starting size
D Allowed error limit

When a computer stores 3.14159 as 3.142, which type of error is introduced?

A Round-off error
B Truncation error
C Residual error
D Relative error

In numerical work, what is the absolute error for exact = 8 and approx = 7.96?

A 0.004
B 0.04
C 0.4
D 4

Relative error is best described as:

A |approx − exact|
B |error| × |exact|
C |exact| − |approx|
D |error| / |exact|

If an iteration reduces error by a constant factor each step, it is called:

A Quadratic convergence
B Cubic convergence
C Linear convergence
D No convergence

In quadratic convergence, the new error is roughly proportional to:

A eₙ
B √eₙ
C 1/eₙ
D eₙ²

In root-finding, the residual at x = xₙ is commonly:

A |f(xₙ)|
B |xₙ|
C |f′(xₙ)|
D |xₙ₊₁ − xₙ|

Which method guarantees the root stays inside the chosen interval every step?

A Newton method
B Bisection method
C Secant method
D Fixed point

Bisection method needs which key requirement for continuous f(x)?

A f′(x) exists
B f(a) = f(b)
C a = 0 always
D f(a)f(b) < 0

Bisection method is most directly supported by:

A Intermediate value theorem
B Taylor theorem
C Cauchy theorem
D Rolle’s theorem

After each bisection step, the interval length becomes:

A Doubled
B Tripled
C Halved
D Unchanged

Bisection method is often described as:

A Fast but risky
B Derivative based
C Matrix based
D Slow but reliable

The false position method is also called:

A Aitken method
B Regula falsi
C Euler method
D Simpson method

False position finds the next estimate using:

A Linear interpolation
B Interval midpoint
C Quadratic approximation
D Random selection

Like bisection, false position requires:

A Two equal endpoints
B Derivative at start
C Complex values
D Sign change bracket

A common drawback of false position is:

A Always diverges
B Needs matrix inverse
C Endpoint stagnation
D Needs second derivative

Newton’s method uses which geometric idea to update the root estimate?

A Tangent line step
B Midpoint rule
C Area under curve
D Parabola fitting

Newton–Raphson iteration formula is:

A xₙ₊₁ = (a+b)/2
B xₙ₊₁ = xₙ + f
C xₙ₊₁ = f′/f
D xₙ₊₁ = xₙ − f/f′

Newton’s method generally needs what additional input compared to secant?

A Sign change bracket
B Derivative values
C Two endpoints
D Matrix diagonal

Newton’s method can fail badly when:

A f′(xₙ) ≈ 0
B f(x) is continuous
C interval is small
D root is real

Near a simple root with good initial guess, Newton often shows:

A Linear convergence
B No convergence
C Quadratic convergence
D Cyclic behavior

Secant method starts with:

A One initial guess
B Three initial guesses
C No starting value
D Two initial guesses

Secant method avoids which requirement of Newton?

A Derivative computation
B Continuity
C Function evaluation
D Root existence

Secant method convergence order is approximately:

A 1.0
B 1.618
C 2.0
D 3.0

Compared to bisection, secant is usually:

A Slower always
B Always divergent
C Always bracketed
D Faster usually

Which root method does NOT require sign-change bracketing?

A Secant method
B Bisection method
C False position
D Both A and B

In fixed-point iteration, the update rule is:

A xₙ₊₁ = xₙ − f/f′
B xₙ₊₁ = (a+b)/2
C xₙ₊₁ = g(xₙ)
D xₙ₊₁ = Ax − b

A common basic convergence condition for fixed-point iteration is:

A |g′(x)| > 1
B g′(x) = 0 always
C f(x) = 1 always
D |g′(x)| < 1

Gauss–Jacobi method is mainly used to solve:

A Linear systems
B Integrals
C Curve tracing
D Differential roots

In Jacobi method, the new values use:

A Only updated values
B Only old values
C Random values
D Exact inverse

A common sufficient condition for Jacobi convergence is:

A Zero diagonal entries
B Non-square matrix
C Negative determinant
D Strict diagonal dominance

Gauss–Seidel differs from Jacobi because it:

A Uses updated values
B Uses midpoint
C Uses no equations
D Uses bracketing

For many diagonally dominant systems, Gauss–Seidel is typically:

A Slower than Jacobi
B Same as bisection
C Faster than Jacobi
D Not iterative

In linear systems, the residual vector is:

A Ax + b
B A − x
C x − b
D b − Ax

Spectral radius of an iteration matrix means:

A Sum of eigenvalues
B Max |eigenvalue|
C Product of eigenvalues
D Number of pivots

A common reason iterative solvers diverge is:

A Spectral radius ≥ 1
B Very small tolerance
C Too many digits
D Continuous function

“Error propagation” in computations mainly means:

A Error always disappears
B Only truncation occurs
C Only rounding occurs
D Error carries forward

Condition number mainly indicates:

A Root location
B Sensitivity measure
C Matrix symmetry
D Exact convergence

A direct method for Ax=b is:

A Gauss elimination
B Jacobi method
C Gauss–Seidel
D Fixed point

LU decomposition is mainly a technique to:

A Find derivatives
B Bracket a root
C Integrate numerically
D Factor a matrix

In bisection, if f(a) and f(mid) have opposite signs, the next interval is:

A [mid, b]
B [a, b] unchanged
C [a, mid]
D [0, mid]

In bisection, midpoint is computed as:

A (a + b)/2
B a − b
C ab/2
D a/b

Newton’s method is faster than bisection mainly because it uses:

A Random intervals
B Only integers
C Matrix splitting
D Slope information

Secant method uses slope approximated from:

A One point only
B Two recent points
C Three midpoints
D Exact derivative

A standard stopping rule in root finding is:

A f(xₙ) maximum
B xₙ negative
C |f(xₙ)| small
D interval grows

A standard stopping rule in iterations can also be:

A |xₙ₊₁ − xₙ| large
B f′(xₙ) large
C a = b
D |xₙ₊₁ − xₙ| small

Aitken’s Δ² method is mainly used to:

A Speed up convergence
B Compute integrals
C Bracket roots
D Solve determinants

SOR is an extension of:

A Newton method
B Bisection method
C Secant method
D Gauss–Seidel method

“Truncation error” is most closely linked to:

A Limited digits storage
B Replacing infinite by finite
C Using wrong bracket
D Derivative becoming zero

“Stability of iteration” in simple terms refers to:

A Small errors not explode
B Always same root
C Using large step
D No function needed

Leave a Reply

Your email address will not be published. Required fields are marked *