Competing Function Model Validation MCQ Practice

Question 1 of 10

A Harborline ferry’s passenger count $P$ is recorded every hour after departure. Successive ratios $P(t+1)/P(t)$ are nearly constant and greater than $1$, while first differences are not constant.

$t$ (h) $P(t)$
$0$ $40$
$1$ $60$
$2$ $90$
$3$ $135$
  1. A. A linear model, because the outputs are positive.
  2. B. An exponential model, because successive ratios are roughly constant.
  3. C. A quadratic model, because the first differences grow.
  4. D. A constant model, because the table has four rows.

Question 2 of 10

A Kelvane tank’s volume $V$ (liters) after $t$ minutes shows nearly constant first differences.

$t$ $V$
$0$ $12$
$1$ $17$
$2$ $22$
$3$ $27$
$4$ $32$
  1. A. $V(t)=12\cdot(5/4)^t$
  2. B. $V(t)=5t+12$
  3. C. $V(t)=2t^2+12$
  4. D. $V(t)=12/t$ for $t>0$

Question 3 of 10

A Ridgeline bridge-sensor height $h$ (cm) against horizontal position $x$ (m) has second differences that are nearly constant and nonzero.

$x$ $h$
$0$ $2$
$1$ $5$
$2$ $10$
$3$ $17$
$4$ $26$
  1. A. Linear, because $h$ is increasing.
  2. B. Exponential, because each output is larger than the previous.
  3. C. Quadratic, because second differences are roughly constant.
  4. D. No polynomial model can fit five points.

Question 4 of 10

Context: a Lumenwick culture of bacteria doubles roughly every equal time interval. Which constructed model type matches that story and the table pattern?

$t$ (h) $N(t)$
$0$ $25$
$1$ $50$
$2$ $100$
$3$ $200$
Scatter of bacteria counts with a competing straight line and a steeper exponential curve through the early points.
  1. A. Linear $N(t)=58.3t+25$, because a line can be drawn near the points.
  2. B. Exponential $N(t)=25\cdot 2^t$, matching doubling and constant ratios.
  3. C. Quadratic opening downward, because population cannot grow forever.
  4. D. Constant $N(t)=25$, because the culture starts at $25$.

Question 5 of 10

A Pellegrin drone’s altitude $a$ (m) over time $t$ (min) rises, then falls, returning near the launch height. Competing models are proposed from the table.

$t$ $a$
$0$ $0$
$1$ $8$
$2$ $12$
$3$ $12$
$4$ $8$
$5$ $0$
  1. A. An exponential growth model, because early values increase.
  2. B. A linear model through $(0,0)$ and $(5,0)$ only, ignoring the midflight peak.
  3. C. A quadratic model with a maximum midflight, matching the symmetric rise and fall.
  4. D. Any model is equally valid because all pass through $(0,0)$.

Question 6 of 10

Nettlewick café sales $S$ (hundreds of dollars) after $d$ days of a promotion show roughly constant first differences. Which choice correctly constructs a linear model from the data?

$d$ $S$
$0$ $4.0$
$2$ $5.6$
$4$ $7.2$
$6$ $8.8$
  1. A. $S(d)=4\cdot(1.4)^d$
  2. B. $S(d)=0.8d+4$
  3. C. $S(d)=0.4d^2+4$
  4. D. $S(d)=4/d$ for $d>0$

Question 7 of 10

A Cobalt lab fits $y=3x+2$ to a data set. At $x=4$ the observed $y$ is $11$. Residual means observed minus predicted. What is the residual, and what does its sign mean?

  1. A. Residual $3$; the model overestimates the observation.
  2. B. Residual $-3$; the model underestimates the observation.
  3. C. Residual $-3$; the model overestimates the observation.
  4. D. Residual $3$; the model underestimates the observation.

Question 8 of 10

Students fit a linear model to Harborline tide heights and list residuals $e=$ observed $-$ predicted.

$t$ Observed Predicted Residual
$1$ $2.4$ $2.1$ $0.3$
$2$ $2.8$ $3.0$ $-0.2$
$3$ $3.5$ $3.9$ $-0.4$
$4$ $4.6$ $4.8$ $-0.2$
  1. A. At $t=1$ the model overestimates, because the residual is positive.
  2. B. At $t=1$ the model underestimates, because the residual is positive.
  3. C. Every residual being nonzero proves the linear model is invalid in every context.
  4. D. Residuals cannot be used to discuss overestimate versus underestimate.

Question 9 of 10

After fitting a line to Kelvane pump data, a residual plot of $e$ versus $t$ shows a clear U-shaped pattern (residuals positive, then negative, then positive). What does that pattern suggest?

$t$ Residual $e$
$0$ $1.2$
$1$ $-0.4$
$2$ $-1.1$
$3$ $-0.3$
$4$ $1.0$
  1. A. The linear model is excellent because residuals change sign.
  2. B. A curved (often quadratic) competing model may fit better, because residuals are systematically patterned.
  3. C. An exponential model is required whenever any residual is negative.
  4. D. Residuals should be discarded; only predictions matter for validation.

Question 10 of 10

A Ridgeline exponential model predicts $18$ units at hour $6$, but the sensor reads $21$. Which validation statement is correct?

  1. A. Residual $+3$; the model underestimates the reading.
  2. B. Residual $+3$; the model overestimates the reading.
  3. C. Residual $-3$; the model underestimates the reading.
  4. D. Residual $0$; predictions never need residual checks.