Study guide

3.2 Parameter Estimation, Linear Regression, Correlation, and Queuing Models

PRC Civil Engineer Licensure Examination CELE

Track your progress as you read

Finished sections get a check mark.

1 Chapter 1: Examination Overview, RA 544 & Professional Practice

1.1 CELE Structure, Format, Scoring, and 2026 Subject Order 1.2 Republic Act No. 544: The Civil Engineering Law and Board Regulations 1.3 Civil Engineering Ethics, Contracts, and Professional Practice

2 Chapter 2: Engineering Mathematics, Numerical Methods & Physics

2.1 Advanced Algebra, Progressions & Word Problems 2.2 Trigonometry and Analytical Geometry 2.3 Differential and Integral Calculus 2.4 Differential Equations and Numerical Methods 2.5 Series, Hyperbolic Functions, Multivariable Calculus, and Laplace Transforms 2.6 Numerical Methods: Linear Systems, Interpolation, Curve Fitting, and ODE Solvers 2.7 Physics for Engineers: Kinematics, Newton's Laws, Work-Energy, and Momentum

3 Chapter 3: Engineering Data Analysis & Engineering Economy

3.1 Engineering Data Analysis and Statistical Applications 3.2 Parameter Estimation, Linear Regression, Correlation, and Queuing Models 3.3 Engineering Economy: Time Value of Money and Interest Formulas 3.4 Project Evaluation, Depreciation, and Benefit-Cost Analysis

4 Chapter 4: Surveying Fundamentals, Construction Layout & Route Surveying

4.1 Errors, Corrections, and Pacing/Taping 4.2 Leveling, Earthworks, and Mass Haul Diagrams 4.3 Traverse Computations, Balancing, and Area Determinations 4.4 Route Surveying: Horizontal and Vertical Curves

5 Chapter 5: Highway & Transportation Engineering

5.1 Highway Geometric Design and Sight Distances 5.2 Traffic Engineering, Flow Fundamentals, and Pavement Design 5.3 Highway Policy, Route Location, Drainage, and Earthwork Slopes 5.4 Transportation Planning, Network Analysis, and Intermodal Systems

6 Chapter 6: Construction Materials, Methods, Management & Safety

6.1 Materials for Construction: Concrete, Formwork, Masonry, Steel, and Wood Products 6.2 Construction Equipment, Formwork, and Construction Methods 6.3 Construction Occupational Safety and Health: RA 11058 and DOLE DO 13 6.4 Construction Project Management, CPM/PERT, and Resource Scheduling 6.5 Construction Principles: Design Requirements, Project Organization, Cost Control, and Value Engineering 6.6 Quantity Surveying, Unit Cost Analysis, and Project Cost Estimates 6.7 Construction Trades and Specialties: Plumbing, Plastics, Doors and Windows, Finishes, and Electrical

7 Chapter 7: Fluid Properties, Hydrostatics & Buoyancy

7.1 Fluid Properties, Units, and Pressure Measurement 7.2 Hydrostatic Pressure on Plane and Curved Surfaces 7.3 Buoyancy, Flotation, and Accelerated Fluid Masses

8 Chapter 8: Hydrodynamics, Pipe Flow, Open Channels & Water Supply

8.1 Fluid Dynamics: Continuity and Energy Equations 8.2 Flow Measurement: Orifices, Weirs, and Venturi Meters 8.3 Fluid Flow in Pipes, Friction Losses, and Pipe Networks 8.4 Open Channel Flow, Manning's Equation, and Hydraulic Jump 8.5 Water Supply: Demand, Storage, Pumping, Distribution, and Treatment

9 Chapter 9: Soil Properties, Seepage & Stresses in Soil Mass

9.1 Soil Composition, Weight-Volume Phase Relationships 9.2 Soil Classification Systems (USCS & AASHTO) and Atterberg Limits 9.3 Soil Permeability, Seepage, Flow Nets, and Effective Stress 9.4 Stresses in Soil Mass, Stress Distribution, and Immediate and Secondary Settlement

10 Chapter 10: Consolidation, Soil Strength, Earth Pressures & Foundations

10.1 Soil Compaction and Soil Compressibility/Consolidation 10.2 Shear Strength of Soils and Mohr-Coulomb Failure Criteria 10.3 Lateral Earth Pressures (Rankine & Coulomb) and Retaining Walls 10.4 Shallow and Deep Foundation Bearing Capacity and Slope Stability

11 Chapter 11: Engineering Mechanics & Statics

11.1 Force Systems, Resultants, and Equilibrium of Rigid Bodies 11.2 Centroids, Moments of Inertia, and Friction 11.3 Cables, Trusses, and Internal Forces in Members

12 Chapter 12: Strength of Materials

12.1 Axial Stress, Strain, and Deformation 12.2 Torsion, Flanged Bolt Couplings, and Thin-Walled Pressure Vessels 12.3 Beam Stresses: Flexural and Transverse Shear 12.4 Beam Deflections, Combined Stresses, and Mohr's Circle

13 Chapter 13: Structural Analysis & NSCP Load Criteria

13.1 Indeterminate Structures: Approximate and Classical Methods 13.2 Moment Distribution Method and Slope Deflection 13.3 Influence Lines for Determinate and Indeterminate Structures 13.4 NSCP Minimum Design Loads: Dead, Live, Wind, and Earthquake

14 Chapter 14: Reinforced Concrete Design (NSCP 2015 / USD)

14.1 Ultimate Strength Design (USD) Principles and Singly Reinforced Beams 14.2 Doubly Reinforced Beams and T-Beams 14.3 Shear and Diagonal Tension, Stirrup Design, and Development Length 14.4 Reinforced Concrete Columns and Isolated Footing Design

15 Chapter 15: Steel, Timber & Prestressed Design and Materials Testing

15.1 Structural Steel Tension Members and Connections 15.2 Structural Steel Compression Members and Beam Design 15.3 Timber Design Principles, Working Stress, and Sawn Lumber Beams 15.4 Prestressed Concrete Fundamentals and Prestress Losses 15.5 Construction Materials Testing and Concrete Mix Proportioning

3.2 Parameter Estimation, Linear Regression, Correlation, and Queuing Models

Key Takeaways Explain This Topic AI

The least-squares slope is b = [nΣxy − ΣxΣy] / [nΣx² − (Σx)²], and the fitted line always passes through (x̄, ȳ).

The coefficient of determination R² is the fraction of the variation in y explained by the regression line.

A correlation coefficient near ±1 shows a strong linear relationship, but correlation alone does not prove causation.

In an M/M/1 queue with arrival rate λ and service rate μ, utilization is ρ = λ/μ and must be below 1 for a stable queue.

Little's law L = λW links the average number in a system to the arrival rate and the average time in the system.

Last updated: October 2026

3.2 Parameter Estimation, Linear Regression, Correlation, and Queuing Models

The AMSTHC TOS area "Engineering Data Analysis" has five competencies:

Describe data and data sets.

Identify important discrete and continuous distributions.

Estimate parameters.

Interpret a linear regression equation.

Apply stochastic processes and queuing models.

Section 3.1 covers descriptive statistics, the main distributions and confidence intervals. This section completes the area.

Estimating Parameters

A parameter describes a population, such as the true mean strength μ \mu μ of a concrete supply. A statistic computed from a sample estimates it.

  • Parameter Point estimator Property
  • Mean μ \mu μ Sample mean x ˉ \bar{x} x ˉ Unbiased

Variance σ 2 \sigma^2 σ 2 s 2 = ∑ ( x i − x ˉ ) 2 / ( n − 1 ) s^2 = \sum (x_i - \bar{x})^2/(n-1) s 2 = ∑ ( x i ​ − x ˉ ) 2 / ( n − 1 ) Unbiased because of the n − 1 n - 1 n − 1 divisor

  • Proportion p p p p ^ = x / n \hat{p} = x/n p ^ ​ = x / n Unbiased
  • Poisson rate λ \lambda λ Observed count ÷ observation time Maximum-likelihood estimate

An interval estimate adds a margin of error, x ˉ ± t α / 2 , n − 1 s / n \bar{x} \pm t_{\alpha/2,\,n-1}\, s/\sqrt{n} x ˉ ± t α /2 , n − 1 ​ s / n ​ . The sample size needed to estimate a mean within ± E \pm E ± E at confidence 1 − α 1 - \alpha 1 − α is:

n = ( z α / 2 σ E ) 2 n = \left(\frac{z_{\alpha/2}\,\sigma}{E}\right)^2 n = ( E z α /2 ​ σ ​ ) 2

Example. Estimate mean daily traffic within ± 200 \pm 200 ± 200 vehicles at 95% confidence when σ ≈ 1,000 \sigma \approx 1{,}000 σ ≈ 1 , 000 . Then n = ( 1.96 × 1,000 / 200 ) 2 = 96.04 n = (1.96 \times 1{,}000/200)^2 = 96.04 n = ( 1.96 × 1 , 000/200 ) 2 = 96.04 , so count 97 days . Always round up.

Simple Linear Regression

To fit y ^ = a + b x \hat{y} = a + bx y ^ ​ = a + b x by least squares , minimize ∑ ( y i − y ^ i ) 2 \sum (y_i - \hat{y}_i)^2 ∑ ( y i ​ − y ^ ​ i ​ ) 2 :

b = n ∑ x y − ∑ x ∑ y n ∑ x 2 − ( ∑ x ) 2 , a = y ˉ − b x ˉ b = \frac{n\sum xy - \sum x \sum y}{n\sum x^2 - \left(\sum x\right)^2}, \qquad a = \bar{y} - b\bar{x} b = n ∑ x 2 − ( ∑ x ) 2 n ∑ x y − ∑ x ∑ y ​ , a = y ˉ ​ − b x ˉ

Because a = y ˉ − b x ˉ a = \bar{y} - b\bar{x} a = y ˉ ​ − b x ˉ , the line always passes through the point ( x ˉ , y ˉ ) (\bar{x}, \bar{y}) ( x ˉ , y ˉ ​ ) .

Correlation coefficient:

r = n ∑ x y − ∑ x ∑ y [ n ∑ x 2 − ( ∑ x ) 2 ] [ n ∑ y 2 − ( ∑ y ) 2 ] , − 1 ≤ r ≤ 1 r = \frac{n\sum xy - \sum x \sum y}{\sqrt{\left[n\sum x^2 - (\sum x)^2\right]\left[n\sum y^2 - (\sum y)^2\right]}}, \qquad -1 \le r \le 1 r = [ n ∑ x 2 − ( ∑ x ) 2 ] [ n ∑ y 2 − ( ∑ y ) 2 ] ​ n ∑ x y − ∑ x ∑ y ​ , − 1 ≤ r ≤ 1

Coefficient of determination: R 2 = r 2 R^2 = r^2 R 2 = r 2 for simple linear regression. It is the fraction of the variation in y y y explained by x x x . The standard error of estimate is s e = ∑ ( y i − y ^ i ) 2 / ( n − 2 ) s_e = \sqrt{\sum (y_i - \hat{y}_i)^2/(n - 2)} s e ​ = ∑ ( y i ​ − y ^ ​ i ​ ) 2 / ( n − 2 ) ​ .

Worked example: strength versus curing age

Five concrete cylinder pairs give:

  • x x x (age, days) 3 7 14 21 28
  • y y y (strength, MPa) 12 18 23 26 28

Sums: ∑ x = 73 \sum x = 73 ∑ x = 73 , ∑ y = 107 \sum y = 107 ∑ y = 107 , ∑ x y = 1,814 \sum xy = 1{,}814 ∑ x y = 1 , 814 , ∑ x 2 = 1,479 \sum x^2 = 1{,}479 ∑ x 2 = 1 , 479 , ∑ y 2 = 2,457 \sum y^2 = 2{,}457 ∑ y 2 = 2 , 457 , with n = 5 n = 5 n = 5 , x ˉ = 14.6 \bar{x} = 14.6 x ˉ = 14.6 and y ˉ = 21.4 \bar{y} = 21.4 y ˉ ​ = 21.4 .

b = 5 ( 1,814 ) − 73 ( 107 ) 5 ( 1,479 ) − 73 2 = 9,070 − 7,811 7,395 − 5,329 = 1,259 2,066 = 0.609 MPa/day b = \frac{5(1{,}814) - 73(107)}{5(1{,}479) - 73^2} = \frac{9{,}070 - 7{,}811}{7{,}395 - 5{,}329} = \frac{1{,}259}{2{,}066} = 0.609\text{ MPa/day} b = 5 ( 1 , 479 ) − 7 3 2 5 ( 1 , 814 ) − 73 ( 107 ) ​ = 7 , 395 − 5 , 329 9 , 070 − 7 , 811 ​ = 2 , 066 1 , 259 ​ = 0.609 MPa/day

a = y ˉ − b x ˉ = 21.4 − 0.609 ( 14.6 ) = 12.50 MPa a = \bar{y} - b\bar{x} = 21.4 - 0.609(14.6) = 12.50\text{ MPa} a = y ˉ ​ − b x ˉ = 21.4 − 0.609 ( 14.6 ) = 12.50 MPa

r = 1,259 2,066 [ 5 ( 2,457 ) − 107 2 ] = 1,259 2,066 × 836 = 1,259 1,314.2 = 0.958 , R 2 = 0.918 r = \frac{1{,}259}{\sqrt{2{,}066\,[5(2{,}457) - 107^2]}} = \frac{1{,}259}{\sqrt{2{,}066 \times 836}} = \frac{1{,}259}{1{,}314.2} = 0.958, \qquad R^2 = 0.918 r = 2 , 066 [ 5 ( 2 , 457 ) − 10 7 2 ] ​ 1 , 259 ​ = 2 , 066 × 836 ​ 1 , 259 ​ = 1 , 314.2 1 , 259 ​ = 0.958 , R 2 = 0.918

The line y ^ = 12.50 + 0.609 x \hat{y} = 12.50 + 0.609x y ^ ​ = 12.50 + 0.609 x explains about 92% of the variation in strength. As a check, the predicted strength at x ˉ = 14.6 \bar{x} = 14.6 x ˉ = 14.6 days is 12.50 + 0.609 ( 14.6 ) = 21.4 MPa = y ˉ 12.50 + 0.609(14.6) = 21.4\text{ MPa} = \bar{y} 12.50 + 0.609 ( 14.6 ) = 21.4 MPa = y ˉ ​ .

Tip

A computed correlation outside − 1 ≤ r ≤ 1 -1 \le r \le 1 − 1 ≤ r ≤ 1 always means one of the sums is wrong. Recompute ∑ x y \sum xy ∑ x y and ∑ y 2 \sum y^2 ∑ y 2 first; they are the sums most often mis-keyed.

Warning

Strength gain with age is really curved: it levels off. A straight line fits well over 3 to 28 days but should not be extrapolated to 90 days. Regression describes association within the data range; it does not prove cause or justify extrapolation.

Stochastic Processes and Arrivals

A stochastic process is a sequence of random events in time. Two models dominate engineering applications:

Poisson arrivals. Vehicles reaching a toll plaza, or trucks reaching a batch plant, arrive independently at an average rate λ \lambda λ . The number arriving in time t t t is Poisson: P ( n ) = ( λ t ) n e − λ t n ! P(n) = \dfrac{(\lambda t)^n e^{-\lambda t}}{n!} P ( n ) = n ! ( λ t ) n e − λ t ​ .

Exponential headways. The time between Poisson arrivals is exponential: P ( h ≥ t ) = e − λ t P(h \ge t) = e^{-\lambda t} P ( h ≥ t ) = e − λ t . This is the probability that a gap is long enough for a pedestrian or a merging driver.

Example. At λ = 360 \lambda = 360 λ = 360 vehicles/h = 0.1 = 0.1 = 0.1 veh/s, the probability of a gap of at least 8 s 8\text{ s} 8 s is e − 0.8 = 0.449 e^{-0.8} = 0.449 e − 0.8 = 0.449 .

A Markov chain models systems that move between states with fixed transition probabilities. For example, a pavement section's condition rating moves from good to fair to poor from year to year. Multiplying a state vector by the transition matrix gives the next year's distribution.

Queuing Models (M/M/1)

An M/M/1 queue has Poisson arrivals at rate λ \lambda λ , exponential service at rate μ \mu μ , and one server. With utilization ρ = λ / μ < 1 \rho = \lambda/\mu < 1 ρ = λ / μ < 1 :

  • Measure Formula
  • Probability system is empty P 0 = 1 − ρ P_0 = 1 - \rho P 0 ​ = 1 − ρ
  • Probability of n n n in system P n = ( 1 − ρ ) ρ n P_n = (1 - \rho)\rho^n P n ​ = ( 1 − ρ ) ρ n

Average number in system L = ρ 1 − ρ = λ μ − λ L = \dfrac{\rho}{1 - \rho} = \dfrac{\lambda}{\mu - \lambda} L = 1 − ρ ρ ​ = μ − λ λ ​

  • Average number in queue L q = ρ 2 1 − ρ L_q = \dfrac{\rho^2}{1 - \rho} L q ​ = 1 − ρ ρ 2 ​
  • Average time in system W = 1 μ − λ W = \dfrac{1}{\mu - \lambda} W = μ − λ 1 ​

Average waiting time in queue W q = λ μ ( μ − λ ) W_q = \dfrac{\lambda}{\mu(\mu - \lambda)} W q ​ = μ ( μ − λ ) λ ​

Little's law , L = λ W L = \lambda W L = λW and L q = λ W q L_q = \lambda W_q L q ​ = λ W q ​ , holds for almost any queue in steady state.

Worked example: a toll booth

Cars arrive at λ = 240 \lambda = 240 λ = 240 /h, and one booth serves μ = 300 \mu = 300 μ = 300 /h.

ρ = 0.80 \rho = 0.80 ρ = 0.80 .

L = 0.8 / 0.2 = 4 L = 0.8/0.2 = 4 L = 0.8/0.2 = 4 cars.

L q = 0.64 / 0.2 = 3.2 L_q = 0.64/0.2 = 3.2 L q ​ = 0.64/0.2 = 3.2 cars.

W = 1 / ( 300 − 240 ) = 1 / 60 h = 60 s W = 1/(300 - 240) = 1/60\text{ h} = 60\text{ s} W = 1/ ( 300 − 240 ) = 1/60 h = 60 s .

W q = 240 / ( 300 × 60 ) = 0.0133 h = 48 s W_q = 240/(300 \times 60) = 0.0133\text{ h} = 48\text{ s} W q ​ = 240/ ( 300 × 60 ) = 0.0133 h = 48 s .

Check with Little's law: L = λ W = 240 ( 1 / 60 ) = 4 L = \lambda W = 240(1/60) = 4 L = λW = 240 ( 1/60 ) = 4 .

If arrivals rise to 290 290 290 /h, then ρ = 0.967 \rho = 0.967 ρ = 0.967 and W = 1 / 10 h = 6 W = 1/10\text{ h} = 6 W = 1/10 h = 6 minutes. Delay grows explosively as ρ → 1 \rho \to 1 ρ → 1 , which is why facilities are designed well below capacity.

Loading diagram... M/M/1 Queue Measures