1Degree of closeness among repeated measurements
Precision
2Degree of closeness of a measurement to its true value
Accuracy
3A measure of dispersion among observations
Standard deviation
4Higher precision
Indicated by a smaller standard deviation
5A measure of the uncertainty or reliability of an observation
Probable error
6The degree of doubt associated with a measured or computed value
Uncertainty
7A probability distribution describing two or more random variables together
Joint distribution
8A mathematical representation of survey observations
Observation model
9The mathematical relationship between observations and unknown parameters
Functional model
10A statistical description of uncertainties in survey observations
Stochastic model
11Functional model and stochastic model
Two components of an observation model
12\ell = x + \varepsilon
Observation model relationship
13Random errors, systematic errors, and blunders
Three types of errors in surveying
14Unpredictable errors that cannot be completely eliminated
Random errors
15Consistent errors caused by identifiable factors
Systematic errors
16Mistakes caused by human or procedural errors
Blunders
17An uncorrected prism constant
Example of an uncorrected systematic error in a total station
18Atmospheric refraction
Atmospheric phenomenon that can cause survey errors
19Horizontal distance (HD)
Distance that should be used instead of slope distance for horizontal computations
20An incorrectly interpreted or modeled error
Cause of a systematic error when interpreting observations
21The process of determining the most probable values of unknown parameters from observations
Adjustment
22An adjustment method that minimizes the sum of squared residuals
Least squares
23Additional observations beyond the minimum required
Redundancy
24The number of independent redundant observations available for adjustment
Degrees of freedom
25To obtain unique values, handle redundant observations, account for random errors, improve precision, and determine by-products
Purpose of adjustment
26Standard deviations and covariancesExamples of by-products of adjustment
Standard deviations and covariances
27Events in which the occurrence of one does not affect the probability of the other
Independent events
28P(X \le x, Y \le y) = P(X \le x)P(Y \le y)
Condition for independence using joint probability
29F(x, y) = F(x)F(y)
Condition for independence using distribution functions
30E[XY] = E[X]E[Y] = \mu_x\mu_y
Expected value of the product of independent variables
31A measure of the degree and direction of the relationship between two random variables
Covariance
32\operatorname{Cov}(X, Y) = \sigma_{xy}
Covariance of two random variables
33\sigma_{xy} = E[(X - \mu_x)(Y - \mu_y)]
Formula for covariance using expected values
34A positive relationship between variables
Indicated by positive covariance
35No linear correlation between variables
Indicated by zero covariance
36A negative relationship between variables
Indicated by negative covariance
37NoDoes zero covariance necessarily mean there is no relationship?
No
38A linear relationship between variables
Indicated by non-zero covariance
39A measure of the strength and direction of linear correlation between two variables
Pearson product-moment correlation coefficient
40ho_{xy} = \frac{\sigma_{xy}}{\sigma_x\sigma_y}
Formula for the correlation coefficient
41-1 \le ho_{xy} \le +1
Range of the correlation coefficient
42Perfect positive correlation
Indicated by a correlation coefficient of plus one
43No linear correlation
Indicated by a correlation coefficient of zero
44Perfect negative correlation
Indicated by a correlation coefficient of minus one
45Covariance indicates direction and depends on measurement units; correlation indicates strength and direction and is unitlessDifference between covariance and correlation
Covariance indicates direction and depends on measurement units; correlation indicates strength and direction and is unitless
46A triangular distributionDistribution of the sum of two independent uniformly distributed variables over equal intervals
A triangular distribution
47A normal distributionDistribution of the sum of normally distributed variables
A normal distribution
48CovarianceFifth parameter associated with a joint distribution
Covariance
49The expected product of the deviations of two variables from their means
Covariance in terms of deviations from the means
50A square matrix containing the variances and covariances of random variables
Variance-covariance matrix
51Variances of individual observationsRepresented by diagonal elements of a variance-covariance matrix
Variances of individual observations
52Represented by off-diagonal elements of a variance-covariance matrix
Covariances between different observations
53Why a covariance matrix is symmetric
Because \sigma_{xy} = \sigma_{yx}
54Represented by vector X in covariance propagation
The vector consisting of observables
55Represented by vector \mu
The vector of true or expected values
56Represented by vector \varepsilon
The vector of true errors
57The process of determining how uncertainties in input variables affect uncertainties in computed variables
Covariance propagation
58What happens to random errors during covariance propagation
They cannot be completely corrected and must be propagated
59The expected squared deviation from the mean
Variance
60General linear transformation formula
y = Ax + b
61Represented by x in a linear transformation
The input random vector
62Represented by y in a linear transformation
The output random vector
63Represented by A in a linear transformation
The transformation matrix
64Represented by b in a linear transformation
The constant vector
65General form of a two-variable linear transformation
Y_1 = a_{11}X_1 + a_{12}X_2 + b_1 and Y_2 = a_{21}X_1 + a_{22}X_2 + b_2
66Linear covariance propagation formula
\Sigma_y = A\Sigma_x A^T
67Represented by \Sigma_x
The covariance matrix of input variables
68Represented by \Sigma_y
The covariance matrix of output variables
69Represented by A^T
The transpose of the transformation matrix
70A matrix containing the partial derivatives of output variables with respect to input variables
Jacobian matrix
71Purpose of the Jacobian matrix
To propagate covariance for nonlinear functions
72Nonlinear covariance propagation formula
\Sigma_y = J\Sigma_x J^T
73A function in which output variables are linear combinations of input variables
Linear function
74A function in which output variables are not linear combinations of input variables
Nonlinear function
75Variance of a linear combination of uncorrelated variables
\sigma_y^2 = a_1^2\sigma_1^2 + a_2^2\sigma_2^2 + \cdots + a_n^2\sigma_n^2
76Variance of a nonlinear function of uncorrelated variables
\sigma_y^2 = \left(\frac{\partial y}{\partial x_1}ight)^2\sigma_1^2 + \cdots + \left(\frac{\partial y}{\partial x_n}ight)^2\sigma_n^2
77Assumption used in simplified variance propagation formulas
The input variables are uncorrelated
78What must be included when input variables are correlated
Covariance terms
79\sigma_{x+y}^2 = \sigma_x^2 + 2\sigma_{xy} + \sigma_y^2
Variance of the sum of two random variables
80\sigma_{xy} = 0
Covariance of independent variables
81\sigma_{x+y}^2 = \sigma_x^2 + \sigma_y^2
Variance of the sum of independent variables
82\mu_{x+y} = \mu_x + \mu_y
Mean of the sum of two random variables
83A matrix representing the absolute variances and covariances of observations
Covariance matrix
84A matrix representing the relative variances and covariances of observations
Cofactor matrix
85Relationship between covariance and cofactor matrices
C = kQ
86Represented by C in C = kQ
Covariance matrix
87Represented by Q in C = kQ
Cofactor matrix
88An arbitrary scale constant or variance factor
Represented by k in C = kQ
89Represented by diagonal elements of a cofactor matrix
Cofactor variances
90Represented by off-diagonal elements of a cofactor matrix
Cofactor covariances
91A matrix representing the relative importance or reliability of observations
Weight matrix
92W = Q^{-1} = kC^{-1}
Formula for the weight matrix
93Purpose of the inverse cofactor matrix
Determining observation weights
94Measurements whose covariances are zero
Uncorrelated measurements
95Weight formula for an uncorrelated observation
w_i = \frac{1}{\sigma_i^2}
96Represented by w_i
The weight of the i-th observation
97Represented by s_i^2
The variance of the i-th observation
98Weights in which all observations have equal values
Unity weights
99Variance factor or factor of unit weight
k = s_0^2 = \sigma_0^2
100The prior variance or variance factor of unit weight
Represented by \sigma_0^2
101What a weight numerically represents
The relative precision of an observation
102Indicated by a high weight
Higher precision and a better-quality observation
103How a weight matrix is constructed
Using the standard deviations of observations
104Formula for estimating the variance factor
s_0^2 = \frac{\sum v_i^2}{n - p}
105Condition indicating a well-estimated variance factor
If s_0^2 = k, then it approaches one
106Formula for the scaled standard deviation of an observation
\hat{\sigma}_i = \frac{s_0}{\sqrt{w_i}}
107\hat{\sigma}_{ar{x}} = \frac{s_0}{\sqrt{\sum w_i}}
Formula for the scaled standard deviation of the mean
108An individual observation
Represented by l_i in the variance factor formula
109The weighted mean
Represented by \overline{x} in the variance factor formula
110The number of observations
Represented by n in the variance factor formula
111Weighted mean formula
\overline{x} = \frac{\sum w_i l_i}{\sum w_i}
112The weighted average using the number of sets as weights
Overall mean of independent angle observations with equal precision
113Convert them into seconds or use a common angular reference
What should be done to angle observations before calculating their weighted mean
114The difference in elevation between two benchmarks
Elevation difference
115The path followed when measuring the elevation difference between benchmarks
Leveling route
116The distance traveled during leveling
Route length
117Longer routes have lower weights, while shorter routes have higher weights
How route length affects the weight of a leveling measurement
118w_i = \frac{1}{S_i}
Weight formula for leveling measurements when variance is proportional to route length
119\overline{h} = \frac{\sum w_i h_i}{\sum w_i}
Weighted mean formula for elevation differences
120Values calculated from measured quantities
Derived quantities
121The uncertainty of that quantity
What standard deviation indicates for a calculated quantity
122A method of propagating variances and covariances through multiple stages of computation
Stepwise propagation
123They are generally considered independent or uncorrelated
General principle about field measurements in stepwise propagation
124Correlation
What stepwise propagation can introduce among intermediate quantities
125\Sigma_y = J \Sigma_x J^T
General law of propagation of variance
126Partial derivatives of intermediate or output quantities with respect to input quantities
Contained in matrix J
127The covariance matrix of original measurements
Represented by \Sigma_x in stepwise propagation
128The covariance matrix of derived or intermediate quantities
Represented by \Sigma_y in stepwise propagation
129x = [b, \alpha, amma, \delta, \epsilon]^T
Field measurement vector in the two-step height computation example
130y = [a, c, \delta, \epsilon]^T
Intermediate quantity vector in the two-step height computation example
131h = [h]
Height vector in the two-step height computation example
132a = \frac{b\sin(\alpha)}{\sin(\alpha + eta)}
Sine rule formula for side a in the two-step height computation
133c = \frac{b\sin(amma)}{\sin(amma + eta)}
Sine rule formula for side c in the two-step height computation
134h = \frac{1}{2}[a an(\delta) + c an(\epsilon)]
Height formula in the two-step computation example
13594.232 meters
Calculated height in the two-step computation example
1360.032 meters
Calculated standard deviation of height in the two-step computation example
137The ratio of the number of ways an event can occur to the total number of possible outcomes
Probability
138An outcome or group of outcomes of an experiment
Event
139An event involving the simultaneous occurrence of two or more events
Compound event
140Events in which the occurrence of one does not affect the probability of another
Independent events in probability
141An unpredictable measurement error that may be positive or negative
Random error
142A bell-shaped probability distribution commonly used to describe random measurement errors
Normal distribution
143A range around the mean within which a specified percentage of observations is expected to fall
Probable error in probability theory
144P(E) = \frac{m}{m + n}
Basic probability formula
145P(E) = \frac{m}{m + n}
Probability of an event occurring if it can occur in m ways and fail in n ways
146P( ext{not } E) = \frac{n}{m + n}
Probability of an event failing if it can occur in m ways and fail in n ways
1470 \le P(E) \le 1
Range of probability
148The event is impossible
Indicated by a probability of zero
149The event is certain
Indicated by a probability of one
1501
Sum of the probabilities of all possible outcomes
151By multiplying their individual probabilities
How probabilities of independent compound events are calculated
152P = P_1 \cdot P_2 \cdot P_3 \cdots P_n
Formula for the probability of simultaneous independent events
153\frac{1}{2}
Probability of drawing a red ball from a box containing one red and one white ball
154\frac{1}{4}
Probability of drawing two red balls from two boxes, each containing one red and one white ball
155\frac{1}{4}
Probability of drawing two white balls from two boxes, each containing one red and one white ball
156\frac{1}{2}
Probability of drawing one red and one white ball from two boxes, each containing one red and one white ball
157+1 and -1
Two possible values of a random error with magnitude of one unit
158\frac{1}{2}
Probability of each possible value of a random error with magnitude of one unit
159-2, 0, and +2
Possible combined errors when two measurements each have an error of plus one or minus one
160\frac{1}{4}
Probability of a combined error of minus two from two measurements
161\frac{1}{2}
Probability of a combined error of zero from two measurements
162\frac{1}{4}
Probability of a combined error of plus two from two measurements
163T = 2^n
Formula for the total number of possible combinations of random errors
164The total number of possible combinations
Represented by T in the random error combinations formula
165The number of combined measurements
Represented by n in the random error combinations formula
166It approaches a normal distribution
What happens to the distribution of random errors as the number of combined measurements increases
167A bell-shaped curve representing the distribution of random errors
Normal distribution curve
168A mathematical function describing the distribution of probabilities
Probability density function
169The spread of observations around the mean
What standard deviation measures in a normal distribution
170The central or average value of the distribution
What the mean represents in a normal distribution
171f(x) = \frac{1}{\sigma\sqrt{2\pi}}e^{-\frac{(x-\mu)^2}{2\sigma^2}}
Normal probability density function