Business Calculus Quiz: Choosing Models
19 questions · exam conditions
0:00
Choosing ModelsQuestion 1 of 19

The marketing team for a new mobile app observes that the rate of new downloads per day is directly proportional to the number of active users the app currently has. If N(t)N(t) represents the number of active users at time tt, which type of function would best model N(t)N(t)?

A linear model, because 'proportional' implies a constant, steady rate of growth.
An exponential model, because the condition described is dN/dt=kNdN/dt = kN for some constant k>0k>0.
A logarithmic model, because the growth will naturally slow as the market becomes saturated.
A constant model, because the proportionality constant represents a fixed number of users.
← Back to quizzes

Business Calculus Quiz

Business Calculus Quiz: Choosing Models

Practice Choosing Models in Business Calculus with focused quiz questions that help you check what you know, review explanations, and build confidence with test-style prompts.

What this quiz covers

This quiz focuses on Choosing Models, giving you a quick way to practice the rules, question types, and explanations that matter most for Business Calculus.

How to use this quiz

Try each quiz question before looking at the correct answer. Use the explanations to review missed ideas, then come back to similar questions until the pattern feels familiar.

All questions

Question 1

The marketing team for a new mobile app observes that the rate of new downloads per day is directly proportional to the number of active users the app currently has. If N(t)N(t) represents the number of active users at time tt, which type of function would best model N(t)N(t)?

  1. A linear model, because 'proportional' implies a constant, steady rate of growth.
  2. An exponential model, because the condition described is dN/dt=kNdN/dt = kN for some constant k>0k>0. (correct answer)
  3. A logarithmic model, because the growth will naturally slow as the market becomes saturated.
  4. A constant model, because the proportionality constant represents a fixed number of users.
Explanation: The statement 'the rate of new downloads per day is directly proportional to the number of active users' translates mathematically to the differential equation dN/dt=kNdN/dt = kN, where kk is the constant of proportionality. The solution to this differential equation is the exponential function N(t)=N0ektN(t) = N_0 e^{kt}, where N0N_0 is the initial number of users. Therefore, an exponential model is the most appropriate.

Question 2

A factory's total production cost C(q)C(q) for producing qq units is increasing. Due to efficiencies of scale, the cost to produce each additional unit is slightly less than the cost to produce the one before it. Which of the following mathematical conditions best describes this situation and what model does it suggest?

  1. C′(q)>0C'(q) > 0 and C′′(q)>0C''(q) > 0, suggesting an exponential model.
  2. C′(q)>0C'(q) > 0 and C′′(q)<0C''(q) < 0, suggesting a logarithmic model. (correct answer)
  3. C′(q)>0C'(q) > 0 and C′′(q)=0C''(q) = 0, suggesting a linear model.
  4. C′(q)<0C'(q) < 0 and C′′(q)<0C''(q) < 0, suggesting total cost is decreasing.
Explanation: The total cost is increasing, which means its rate of change, the first derivative C′(q)C'(q), must be positive. 'The cost to produce each additional unit' is the marginal cost, which is C′(q)C'(q). The problem states that this marginal cost is decreasing. This means the rate of change of the marginal cost, which is the second derivative C′′(q)C''(q), must be negative. The conditions C′(q)>0C'(q) > 0 and C′′(q)<0C''(q) < 0 describe a function that is increasing but concave down. Among the standard choices, the logarithmic model fits this description.

Question 3

The value of a company's computer hardware is known to depreciate over time. The accounting department's policy is to model the value by assuming the hardware loses 20% of its remaining value each year. If V(t)V(t) is the value of the hardware tt years after purchase, which model is most suitable for V(t)V(t)?

  1. A linear model, because the value decreases by a constant percentage each year.
  2. A logarithmic model, because the value must decrease over time and approach zero.
  3. An exponential model, because the value is multiplied by a constant factor each year. (correct answer)
  4. A linear model, because this represents a constant dollar amount of depreciation annually.
Explanation: Losing 20% of the remaining value each year means that 80% of the value remains. Thus, each year the value is multiplied by a constant factor of 0.8. If the initial value is V0V_0, the value after tt years is V(t)=V0(0.8)tV(t) = V_0(0.8)^t. This is the form of an exponential decay function, which is a type of exponential model. A linear model would imply the value decreases by the same fixed dollar amount each year, not a percentage of the remaining value.

Question 4

At the beginning of a promotional campaign, a website had 500,000 registered users. In the first month, it gained 50,000 new users. In the second month, it gained 60,000 new users. In the third month, it gained 72,000 new users. If this pattern of growth in new users continues, which model best describes the total number of users, UU, as a function of the month, mm?

  1. An exponential model, because the rate of growth is increasing by a constant percentage. (correct answer)
  2. A linear model, because the number of users is increasing each month.
  3. A logarithmic model, because the initial growth is fast but must eventually slow down.
  4. A quadratic model, because the growth rate itself is increasing linearly.
Explanation: When analyzing growth patterns in business contexts, you need to examine how the rate of change itself is changing. Look at what's happening to the monthly gains: 50,000 → 60,000 → 72,000 new users. The key insight is recognizing that each month's gain is 20% larger than the previous month's gain. From 50,000 to 60,000 represents a 20% increase (50,000×1.2=60,00050,000 \times 1.2 = 60,000), and from 60,000 to 72,000 is also a 20% increase (60,000×1.2=72,00060,000 \times 1.2 = 72,000). When the rate of growth increases by a constant percentage, this signals exponential behavior in the underlying function. Answer A correctly identifies this exponential pattern. The growth rate itself is growing exponentially, which means the total user count will follow an exponential model. Answer B is wrong because linear models have constant rates of change. Here, the monthly gains are increasing, not staying constant. Answer C incorrectly suggests logarithmic growth. While logarithmic functions do start fast and slow down, the data shows accelerating growth, not decelerating growth. Answer D might seem tempting since the gains are increasing, but quadratic models show linear increases in the growth rate (adding the same amount each period). Here, the growth rate increases by a constant percentage, not a constant amount. Study tip: To distinguish between exponential and quadratic growth patterns, check whether the rate of change increases by a constant amount (quadratic) or by a constant percentage (exponential). Calculate the ratio between consecutive changes to spot exponential patterns.

Question 5

Two startups, Firm A and Firm B, have revenues that are growing over time, tt. Firm A's revenue is modeled by RA(t)=100t+500R_A(t) = 100t + 500. Firm B's revenue is modeled by RB(t)=500(1.2)tR_B(t) = 500(1.2)^t. Both models are in thousands of dollars. For which forecasting task would the linear model likely be a more reasonable choice than the exponential model, despite the exponential model's higher growth rate?

  1. Forecasting revenue for the next quarter, as both models provide similar short-term values and linear modeling is simpler. (correct answer)
  2. Forecasting revenue 10 years into the future, as exponential growth becomes unrealistic over long periods.
  3. Forecasting revenue when expecting sustained high growth rates, as linear models handle rapid expansion better.
  4. Forecasting revenue in an expanding market where the firm is gaining significant market share over time.
Explanation: When comparing linear and exponential models for business forecasting, you need to consider both the time horizon and the realistic sustainability of different growth patterns. Let's examine what happens with these models over different time periods. In the short term (say, 3 months or 0.25 years), Firm A generates RA(0.25)=100(0.25)+500=525R_A(0.25) = 100(0.25) + 500 = 525 thousand, while Firm B generates RB(0.25)=500(1.2)0.25≈524R_B(0.25) = 500(1.2)^{0.25} ≈ 524 thousand. The values are nearly identical, making the simpler linear model just as useful for immediate forecasting needs. Answer A is correct because when two models yield similar short-term predictions, choosing the simpler linear model makes practical sense. It's easier to calculate, explain to stakeholders, and implement in business planning. Answer B is wrong because exponential models often are appropriate for long-term forecasting when sustained percentage growth is realistic—think compound interest or technology adoption curves. Answer C misunderstands model behavior. Exponential models, not linear ones, are specifically designed to capture rapid, accelerating growth. Linear models show constant growth rates. Answer D describes exactly when exponential models excel. Market expansion and growing market share typically follow exponential patterns as success builds upon itself. Study tip: For model selection questions, always consider the time horizon and growth sustainability. Linear models work well for short-term forecasting and situations with constant absolute changes, while exponential models suit scenarios with constant percentage growth rates.

Question 6

A manufacturing plant's output PP (in units per hour) is a function of the number of workers NN. Initially, adding more workers leads to a significant increase in output. However, due to limited space and resources, as more workers are added, each additional worker contributes less to the total output than the worker before them, although the total output continues to increase. Which model best describes P(N)P(N)?

  1. A linear model, as output increases with the number of workers.
  2. An exponential model, as the output grows rapidly with more workers.
  3. A linear model with a negative slope, because each worker is less effective.
  4. A logarithmic model, as it represents production with diminishing marginal returns. (correct answer)
Explanation: When you encounter production scenarios describing diminishing returns, you're dealing with a classic economic concept that translates directly into mathematical modeling. The key phrase here is "each additional worker contributes less...although total output continues to increase" – this describes diminishing marginal returns. A logarithmic model perfectly captures this behavior. The function P(N)=aln⁡(N)+bP(N) = a\ln(N) + b starts steep (rapid initial growth when adding the first workers) but gradually flattens as N increases. The derivative P′(N)=aNP'(N) = \frac{a}{N} shows that the rate of increase (marginal product) decreases as N grows, while the function itself keeps increasing – exactly what the problem describes. Answer A misses the diminishing returns aspect entirely. Linear models have constant marginal returns (each worker adds the same amount), not decreasing ones. Answer B suggests exponential growth, which would mean accelerating returns – the opposite of what's described. Exponential models show increasing marginal returns, where each additional worker becomes more productive. Answer C combines two errors: it's both linear (missing diminishing returns) and suggests decreasing total output, contradicting the problem statement that "total output continues to increase." For business calculus problems involving production, remember this pattern: if marginal returns are diminishing but total output keeps rising, think logarithmic. If you see "each additional unit contributes less than the previous one," logarithmic models are often your answer. This concept appears frequently in economics-based calculus applications.

Question 7

An initial principal of $10,000 is invested in one of two funds. Fund A uses simple interest, adding $500 to the principal at the end of each year. Fund B uses compound interest, increasing the value by 5% of the current balance at the end of each year. Which models best describe the value $Voftheinvestmentasafunctionoftimeof the investment as a function of timet$ in years for Fund A and Fund B, respectively?

  1. Fund A: Exponential; Fund B: Linear
  2. Fund A: Exponential; Fund B: Exponential
  3. Fund A: Linear; Fund B: Linear
  4. Fund A: Linear; Fund B: Exponential (correct answer)
Explanation: When you encounter investment problems, the key is identifying whether the growth pattern is additive (linear) or multiplicative (exponential). Look at how the interest is calculated and applied. Fund A uses simple interest, adding a fixed $500 each year regardless of the current balance. This creates the model $V=10,000+500tV = 10,000 + 500t $, where the value increases by the same amount annually. This is a linear function with a constant rate of change. Fund B uses compound interest, multiplying the current balance by 1.05 each year. This creates the model V = 10,000(1.05)^t , where the dollar amount of growth increases each year as the base grows. This is an exponential function where the rate of change depends on the current value. Choice A incorrectly reverses the models, suggesting Fund A is exponential and Fund B is linear. This represents a fundamental misunderstanding of how simple versus compound interest works. Choice B claims both funds are exponential. While Fund B is correctly identified as exponential, Fund A's fixed annual addition creates linear growth, not exponential. Choice C suggests both funds are linear. This misses that Fund B's percentage-based growth compounds, creating accelerating returns rather than constant additions. Remember this pattern: simple interest (fixed dollar amounts added) always produces linear models, while compound interest (percentage of current balance) always produces exponential models. The words "simple" and "compound" are your immediate clues to distinguish between linear and exponential growth in finance problems.

Question 8

An analyst is modeling a company's profit, P(t)P(t), in millions of dollars, where tt is the number of years since 2010. The analyst discovers that a plot of ln⁡(P(t))\ln(P(t)) versus tt forms an almost perfect straight line with a positive slope. What does this finding imply about the most suitable model for P(t)P(t)?

  1. P(t)P(t) is best modeled by a linear function of tt.
  2. P(t)P(t) is best modeled by a logarithmic function of tt.
  3. P(t)P(t) is best modeled by an exponential function of tt. (correct answer)
  4. P(t)P(t) is best modeled by a power function of tt.
Explanation: If a plot of ln⁡(P(t))\ln(P(t)) versus tt is a straight line, it means there is a linear relationship between these two variables. We can write this relationship as ln⁡(P(t))=mt+b\ln(P(t)) = mt + b for some constants mm and bb. To solve for P(t)P(t), we exponentiate both sides: eln⁡(P(t))=emt+be^{\ln(P(t))} = e^{mt+b}. This simplifies to P(t)=emtebP(t) = e^{mt}e^b. Since ebe^b is just another constant (let's call it AA) and emt=(em)te^{mt} = (e^m)^t, we have P(t)=A(k)tP(t) = A(k)^t where k=emk=e^m. This is the general form of an exponential function.

Question 9

A company finds that its total sales revenue, SS, is a function of its advertising expenditure, AA. An analysis of sales data indicates that as advertising spending increases, sales revenue also increases, but each subsequent $1,000 increase in advertising expenditure results in a progressively smaller increase in sales revenue. Which model best describes the relationship $S(A)$?

  1. A linear model, because sales revenue is consistently increasing with advertising expenditure.
  2. An exponential model, because this represents a situation of accelerating growth in sales.
  3. A logarithmic model, because it captures the principle of diminishing marginal returns. (correct answer)
  4. An exponential decay model, because the effect of advertising is diminishing over time.
Explanation: The scenario describes diminishing marginal returns: sales SS increase with advertising AA, so the first derivative S′(A)S'(A) is positive. However, the increase becomes smaller for each additional dollar spent, meaning the rate of increase is decreasing. This implies the second derivative S′′(A)S''(A) is negative. A function with a positive first derivative and a negative second derivative is characteristic of a logarithmic model (or other concave down functions like a square root function), making it the best choice among the given options.

Question 10

The productivity of a new employee, measured by the number of units UU assembled per day, is found to be a function of the number of weeks of experience, ww. A plot of UU versus ln⁡(w)\ln(w) is approximately a straight line. Which of the following statements is the most accurate conclusion?

  1. The relationship between units assembled and weeks of experience is exponential.
  2. The relationship between units assembled and weeks of experience is logarithmic. (correct answer)
  3. The employee's productivity increases by a constant amount each week.
  4. The employee's productivity increases by a constant percentage each week.
Explanation: If a plot of UU versus ln⁡(w)\ln(w) is a straight line, it means that UU is a linear function of ln⁡(w)\ln(w). This can be written as U(w)=mln⁡(w)+bU(w) = m \ln(w) + b for some constants mm and bb. This is, by definition, a logarithmic model. An exponential relationship would mean ln⁡(U)\ln(U) is a linear function of ww. A constant increase per week would imply a linear model, U(w)=mw+bU(w) = mw+b. A constant percentage increase would imply an exponential model.

Question 11

A company tracks its monthly revenue over 8 months and finds the following pattern: Month 1: $10,000, Month 2: $12,000, Month 3: $14,400, Month 4: $17,280. The CEO believes this growth will continue and wants to model future revenue. However, the CFO argues that such rapid growth is unsustainable and expects the growth rate to decrease over time, eventually leveling off at around $50,000 per month. Which combination of models best represents these two perspectives?

  1. CEO: Linear model, CFO: Exponential model with horizontal asymptote at $50,000
  2. CEO: Exponential model, CFO: Logarithmic model with vertical asymptote at $50,000
  3. CEO: Exponential model, CFO: Logistic model with horizontal asymptote at $50,000 (correct answer)
  4. CEO: Logarithmic model, CFO: Linear model with slope approaching zero
Explanation: The revenue data shows exponential growth (each month is 1.2 times the previous), so the CEO's optimistic view aligns with an exponential model. The CFO's perspective of decreasing growth rate that levels off describes a logistic model with carrying capacity at $50,000. Choice A is wrong because the data isn't linear. Choice B is wrong because logarithmic models show decreasing growth rates from the start, not the observed exponential pattern, and vertical asymptotes don't make sense for revenue. Choice D is wrong because the current data doesn't show logarithmic behavior.

Question 12

A financial advisor analyzes investment returns and finds that Portfolio A's value follows VA(t)=10000(1.08)tV_A(t) = 10000(1.08)^t while Portfolio B's value follows VB(t)=8000+1200tV_B(t) = 8000 + 1200t, where tt is time in years. A client asks when Portfolio A will be worth exactly twice as much as Portfolio B. To solve this problem, the advisor needs to understand what type of equation will result from setting VA(t)=2VB(t)V_A(t) = 2V_B(t).

  1. A linear equation that can be solved algebraically using standard techniques
  2. A quadratic equation requiring the quadratic formula or factoring methods
  3. A transcendental equation requiring logarithmic methods or numerical approximation (correct answer)
  4. A rational equation that simplifies to a polynomial after clearing denominators
Explanation: Setting 10000(1.08)t=2(8000+1200t)10000(1.08)^t = 2(8000 + 1200t) results in 10000(1.08)t=16000+2400t10000(1.08)^t = 16000 + 2400t, which contains both exponential and linear terms. This is a transcendental equation that cannot be solved by elementary algebraic methods and typically requires logarithmic techniques or numerical methods. Choice A is wrong because exponential terms don't reduce to linear form. Choice B is wrong because there are no squared terms. Choice D is wrong because there are no rational expressions that would clear to a polynomial.

Question 13

A tech company launches a new app and tracks daily downloads. The product manager notices that the relationship between days since launch (x) and total downloads (y) shows specific characteristics: the rate of new downloads is highest on day 1 and decreases over time, but the total downloads continue to increase toward what appears to be a maximum adoption level.

Given the scenario described above, which mathematical model would best represent the total downloads as a function of days since launch?

  1. y=L(1−e−kx)y = L(1 - e^{-kx}) where L>0L > 0 and k>0k > 0, representing limited growth (correct answer)
  2. y=aln⁡(x+1)+by = a\ln(x + 1) + b where a>0a > 0, representing logarithmic growth
  3. y=ae−bxy = ae^{-bx} where a>0a > 0 and b>0b > 0, representing exponential decay
  4. y=ax2+bx+cy = ax^2 + bx + c where a<0a < 0, representing quadratic growth with maximum
Explanation: When analyzing growth patterns in business applications, you need to match the described behavior with the mathematical properties of different function types. This scenario describes three key characteristics: downloads start at zero, grow rapidly initially, then approach a maximum level as the growth rate decreases over time. Option A, y=L(1−e−kx)y = L(1 - e^{-kx}), perfectly captures this behavior. This limited growth model starts at zero when x=0x = 0, since y=L(1−e0)=L(1−1)=0y = L(1 - e^0) = L(1 - 1) = 0. As time increases, e−kxe^{-kx} approaches zero, so yy approaches the limiting value LL. The derivative y′=Lke−kxy' = Lke^{-kx} is highest at x=0x = 0 and decreases over time, matching the described download pattern. Option B, logarithmic growth y=aln⁡(x+1)+by = a\ln(x + 1) + b, grows without bound and never reaches a maximum adoption level, contradicting the scenario's mention of approaching a maximum. Option C, exponential decay y=ae−bxy = ae^{-bx}, starts at its maximum value aa and decreases toward zero, which is the opposite of what's described. Option D, the quadratic model y=ax2+bx+cy = ax^2 + bx + c with a<0a < 0, does have a maximum, but after reaching it, downloads would decrease, which doesn't match real-world app adoption patterns. Study tip: Limited growth models (also called logistic growth) frequently appear in business calculus because they model realistic scenarios where growth has natural constraints—market saturation, resource limits, or adoption ceilings. Always look for the S-shaped curve behavior when growth starts slow, accelerates, then levels off.

Question 14

An economist studies the relationship between years of education (xx) and lifetime earnings (yy, in thousands). Three proposed models are: Model 1: y=200+50xy = 200 + 50x, Model 2: y=150⋅1.12xy = 150 \cdot 1.12^x, Model 3: y=100+180ln⁡(x+1)y = 100 + 180\ln(x + 1). If research shows that each additional year of education provides significant benefit, but the marginal benefit of education decreases as education level increases, and there's no theoretical upper limit to earnings, which model is most appropriate?

  1. Model 1, because it shows consistent returns to education investment over time
  2. Model 2, because it captures the compounding effect of education on earning potential
  3. Model 2, because exponential models always show diminishing marginal returns
  4. Model 3, because it shows diminishing marginal returns without an upper bound (correct answer)
Explanation: When analyzing mathematical models for economic relationships, you need to examine both the marginal behavior (how the rate of change evolves) and any upper bounds the model imposes. The research indicates two key properties: diminishing marginal returns (each additional year of education helps less than the previous year) and no theoretical upper limit to earnings. This means you need a function where the derivative decreases as xx increases, but the function itself can grow indefinitely. Model 3, y=100+180ln⁡(x+1)y = 100 + 180\ln(x + 1), perfectly fits these criteria. Taking the derivative: dydx=180x+1\frac{dy}{dx} = \frac{180}{x + 1}. This derivative is always positive (education always helps) but decreases as xx increases (diminishing marginal returns). The logarithmic function grows without bound, satisfying the "no upper limit" requirement. Answer A is incorrect because Model 1 has a constant derivative of 50, meaning each year of education provides exactly the same benefit—contradicting the diminishing returns finding. Answer B correctly identifies that Model 2 grows without bound, but exponential functions actually show increasing marginal returns, not decreasing ones. Answer C contains a fundamental misconception—exponential models show increasing marginal returns, not diminishing ones. The derivative of 150⋅1.12x150 \cdot 1.12^x grows larger as xx increases. Remember: logarithmic functions are your go-to models when you need diminishing marginal returns with unlimited growth potential. Linear models give constant returns, while exponential models give increasing returns.

Question 15

A retail analyst studies three different product categories over a 6-month period. Category A shows month-over-month growth rates of 8%, 8%, 8%, 8%, 8%. Category B shows monthly sales increases of $1000, $1000, $1000, $1000, $1000. Category C shows monthly increases that started at $2000 in month 1, then $1600, $1280, $1024, $819.20. The analyst needs to project sales for month 12. Which statement about model selection is most accurate?

  1. All three categories require exponential models since they all show growth patterns
  2. Category A needs exponential, Category B needs linear, Category C needs logarithmic modeling
  3. Categories A and C both need exponential models, while Category B needs polynomial modeling
  4. Category A needs exponential, Category B needs linear, Category C's increments follow exponential decay (correct answer)
Explanation: When you encounter growth pattern analysis in business calculus, you need to identify the underlying mathematical relationship by examining how values change over time. Look at the pattern of change itself, not just whether numbers are increasing. Category A shows constant 8% growth rates, which means each month's sales equal the previous month times 1.08. This multiplicative relationship defines exponential growth: S(t)=S0⋅(1.08)tS(t) = S_0 \cdot (1.08)^t. Category B adds exactly $1000 each month, creating the linear relationship $S(t)=S0+1000tS(t) = S_0 + 1000t $ since the change is additive and constant. Category C is trickier—while sales are increasing, examine the increments themselves: $2000, $1600, $1280, $1024, 819.20.Eachincrementis80819.20. Each increment is 80% of the previous increment (1600 = $2000 × 0.8, $1280 = $1600 × 0.8, etc.). The increments follow exponential decay, making the overall sales pattern follow a decaying exponential added to initial conditions. Choice A incorrectly assumes all growth requires exponential models—linear growth is still growth but follows addition, not multiplication. Choice B misidentifies Category C as logarithmic; logarithmic growth shows decreasing rates that approach zero, but the pattern here involves exponentially decaying increments. Choice C suggests polynomial modeling for Category B, but constant additive increases are purely linear, and polynomial models involve variable rates of change. Study tip: Always analyze the pattern of change itself. Constant percentage changes = exponential, constant amount changes = linear, and when increments themselves follow a pattern, identify that pattern first.

Question 16

A startup's user base grows from 1,000 to 2,000 in the first month, then to 4,000 in the second month. The growth continues this pattern for three more months, reaching 32,000 users. However, market research indicates that the total addressable market is only 500,000 users. The founder wants to project user growth for the next 12 months. What modeling approach should be recommended?

  1. Continue with exponential model N(t)=1000⋅2tN(t) = 1000 \cdot 2^t since the pattern has been consistent
  2. Switch to logistic model N(t)=5000001+ae−btN(t) = \frac{500000}{1 + ae^{-bt}} to account for market saturation (correct answer)
  3. Use logarithmic model N(t)=a+bln⁡(t)N(t) = a + b\ln(t) to show realistic slowing growth
  4. Apply linear model N(t)=1000+6200tN(t) = 1000 + 6200t based on average monthly increase
Explanation: While the initial growth follows an exponential pattern (doubling each month), the finite market size of 500,000 users means this growth cannot continue indefinitely. A logistic model accounts for both the current exponential behavior and the eventual saturation at the market limit. Choice A ignores market constraints and would predict unrealistic growth. Choice C (logarithmic) doesn't fit the observed exponential pattern in early stages. Choice D (linear) completely misrepresents the doubling pattern and gives an incorrect average increase.

Question 17

An analyst examines the relationship between advertising spend (xx, in thousands) and sales revenue (yy, in thousands) for a product launch. The data suggests that small increases in advertising have dramatic effects initially, but additional spending yields diminishing returns, with sales appearing to approach a maximum of about $200,000 regardless of further advertising investment. If the analyst needs to choose between $y=50+30ln⁡(x+1)y = 50 + 30\ln(x + 1) ,, y=50(1.15)xy = 50(1.15)^x ,and, and y=200−150e−0.2xy = 200 - 150e^{-0.2x} $, which model is most appropriate and why?

  1. y=50+30ln⁡(x+1)y = 50 + 30\ln(x + 1) because it shows diminishing returns without an upper bound constraint
  2. y=50(1.15)xy = 50(1.15)^x because it captures the dramatic initial effects of advertising spending
  3. y=200−150e−0.2xy = 200 - 150e^{-0.2x} because it shows diminishing returns with sales approaching $200,000 (correct answer)
  4. y=50+30ln⁡(x+1)y = 50 + 30\ln(x + 1) because logarithmic models are always best for diminishing returns scenarios
Explanation: The scenario describes diminishing returns with sales approaching a maximum of $200,000. The model $y=200−150e−0.2xy = 200 - 150e^{-0.2x} isexponentialdecaytowardahorizontalasymptoteatis exponential decay toward a horizontal asymptote at y=200y = 200 $, perfectly matching this description. Choice A is wrong because while logarithmic models show diminishing returns, they don't have an upper bound. Choice B is wrong because exponential growth models don't show diminishing returns. Choice D is wrong because logarithmic models aren't always best for diminishing returns when there's a specified upper limit.

Question 18

A pharmaceutical company studies drug concentration in blood over time. The concentration peaks immediately after injection, then decreases. Two researchers propose different models: Dr. Smith suggests C(t)=100e−0.3tC(t) = 100e^{-0.3t} and Dr. Jones suggests C(t)=120−15tC(t) = 120 - 15t. After 4 hours, the measured concentration is 30 mg/L. After 8 hours, it's 10 mg/L. Based on this information and the nature of drug metabolism, which model is more appropriate and why?

  1. Dr. Smith's model, because drug elimination typically follows exponential decay kinetics (correct answer)
  2. Dr. Jones's model, because it better fits the data points at t = 4 and t = 8
  3. Dr. Smith's model, because it predicts the correct concentration at both time points
  4. Neither model, because drug concentration should follow a logarithmic decline pattern
Explanation: Drug elimination typically follows first-order kinetics, resulting in exponential decay. Dr. Smith's model gives C(4) = 100e^(-1.2) ≈ 30.1 mg/L and C(8) = 100e^(-2.4) ≈ 9.1 mg/L, closely matching the data. Dr. Jones's linear model gives C(4) = 60 mg/L and C(8) = 0 mg/L, which doesn't fit. More importantly, linear decay is not physiologically realistic for drug metabolism. Choice B is factually wrong about data fitting. Choice C is partially correct about data fitting but misses the physiological reasoning. Choice D is wrong because logarithmic decline is not typical for drug metabolism.

Question 19

A data scientist examines three different business metrics over time and needs to choose appropriate models. Metric X grows by adding 50 units each month. Metric Y grows by 15% each month. Metric Z grows rapidly initially but the growth rate slows down, with increases of 100, 80, 64, 51.2, and 41 units in consecutive months. If all three metrics started at 200 units, which model assignments are correct?

  1. Metric X: Linear, Metric Y: Exponential, Metric Z: Logarithmic with adjusted parameters
  2. Metric X: Linear, Metric Y: Exponential, Metric Z: Limited growth model with decreasing increments (correct answer)
  3. Metric X: Linear function, Metric Y: Exponential function, Metric Z: Logarithmic function
  4. Metric X: Linear, Metric Y: Compound growth, Metric Z: Power function with fractional exponent
Explanation: Metric X adds constant amounts (50 units/month) = linear model. Metric Y grows by constant percentage (15%/month) = exponential model. Metric Z's increments (100, 80, 64, 51.2, 41) decrease by 20% each time, showing exponentially decreasing growth increments that approach zero, characteristic of a limited growth model. Choice A is wrong because Metric Z doesn't follow a logarithmic pattern. Choice C is wrong because logarithmic functions don't match Metric Z's pattern. Choice D is wrong because 'compound growth' is just another term for exponential, and the pattern for Z doesn't suggest a power function.