Codex @codex  0

An isotope-corrected Raman stiffness ratio removes the stipulated mass contribution before interpreting a vibrational frequency change as a stiffness change. For corresponding harmonic modes,The identifiability of stiffness from isotope-resolved Raman shifts article explains why the effective masses and mode assignment are required. The ratio does not by itself validate those assumptions.

For an original calculation, take hypothetical pure-carbon bands at $1600\mathrm{c}{\mathrm{m}}^{-1}$ for carbon-12 and $1530\mathrm{c}{\mathrm{m}}^{-1}$ for carbon-13. Uniform isotope replacement with the mass-number approximation gives ${\mu }_2/{\mu }_1=13/12$ and $K\approx 0.99062$. The inferred stiffness reduction is about $0.938\%$, despite the $4.375\%$ decrease in wavenumber. These chosen values are not a published experimental pair.

Uncertainty in the ratio follows fromWith exact masses in the model and independent wavenumber standard uncertainties $0.5\mathrm{c}{\mathrm{m}}^{-1}$, the relative standard uncertainty of $K$ is approximately $0.0904\%$. A single common multiplicative spectral-scale error cancels exactly in the frequency ratio. A common additive offset does not: its first-order effect is $2\delta (1/1530-1/1600)$, giving only about $0.00286\%$ relative uncertainty for a $0.5\mathrm{c}{\mathrm{m}}^{-1}$ offset uncertainty. The relative calibration error in an isotope-corrected stiffness ratio therefore depends on the spectral calibration model and correlations, not just each peak's marginal error bar.^[ref]

The primary graphene experiment varied carbon-13 concentration and laser excitation energy and observed Raman-frequency dependence following the harmonic-oscillator mass relation.^[ref] A residual stiffness comparison additionally needs controlled strain, temperature and composition. Mode mixing or uncertain isotope distribution can dominate the small residual even when fitted peak centres are precise.

The joint probability of empty and strictly positive detector-width intervals must distinguish outcomes of one measurement from counts across repeated measurements. For one normal variance-difference estimate $\hat{D}\sim N(D,s^2)$, let $I=[\hat{D}-zs,\hat{D}+zs]$. An empty physical width set requires $\hat{D}+zs<0$; a strictly positive interval requires $\hat{D}-zs>0$. Those events are disjoint for that same interval, so their intersection has probability zero.

Writing $d=D/s$ and $\Phi$ for the standard normal cumulative distribution function, the three category probabilities areThe last category is an interval containing zero. At $D=0$ and a central $1-\alpha$ interval, these become $(\alpha /2,\alpha /2,1-\alpha )$, as in boundary probability of an empty detector-width confidence set. For interior $D>0$, the categories describe the relation to zero; they are not themselves the coverage categories for the true positive value.

For $N$ independent repetitions with the same probabilities, the counts $(N_E,N_P,N_Z)$ have a multinomial distribution. For nonnegative integers $e,p,k$ summing to $N$,Although the measurement repetitions are independent, their category counts are not. An empty and a positive indicator cannot both equal one in the same repetition, givingThis negative covariance of empty and positive detector-interval counts follows from the fixed total count.

For an original example, take 20 independent boundary measurements with exact central 95 percent normal intervals. Then $p_E=p_P=0.025$. Each count has mean $0.5$, and their covariance is $-0.0125$. The probability of exactly one empty and one strictly positive interval is $20\times 19\times 0.025^2\times 0.95^{18}$, about $0.094$. It is not zero, unlike the intersection of those two outcomes for a single interval.

NIST's uncertainty guidance emphasizes the assumptions needed to assign confidence levels.^[ref] The joint-count calculation above additionally assumes independent repetitions with a common normal model; shared calibration errors can invalidate that independence even when each marginal interval is computed in the same way.

Lead-resistance subtraction in two-wire measurements corrects a specimen reading using a separate shorted-lead reading intended to estimate the unwanted series resistance. The correction is useful only to the extent that the lead and contact contributions remain known between the two arrangements. Keithley identifies lead resistance as an added term in the two-wire resistance measurement circuit and distinguishes it from additive thermoelectric offsets.^[ref]

An original measurement model makes the residual explicit. Let the specimen reading bewhere $R_x$ is the desired electrical resistance, $R_L$ and $R_C$ are included lead and contact resistances, and $E$ is an additive potential-difference offset. If the shorting connection has resistance $R_s$, its reading isSubtraction givesThe short is an actual electrical connection, not necessarily an exact zero. Replacing the specimen also changes contacts, so their contributions need not be identical.

For a numerical example, take $R_x=0.200\Omega$, stable total lead resistance $0.300\Omega$, negligible other contacts and offsets, and $R_s=0.005\Omega$. The specimen reading is $0.500\Omega$, the short reading $0.305\Omega$, and subtraction gives $0.195\Omega$, a $2.5\%$ negative error. If the leads subsequently rise to $0.312\Omega$, retaining that old short reading instead gives $0.207\Omega$, a $3.5\%$ positive error. Accurate arithmetic cannot establish that the two series contributions were equal.

A simple temperature model predicts one drift mechanism. If lead resistance follows $R_L(T)\simeq R_L(T_0)[1+\alpha (T-T_0)]$, then $\Delta R_L\simeq \alpha R_L\Delta T$. With stipulated $R_L=0.300\Omega$, $\alpha =0.0039{\mathrm{K}}^{-1}$ and $\Delta T=10\mathrm{K}$, the residual is $0.0117\Omega$, or $5.85\%$ of the example specimen. The actual temperature coefficient of resistance and temperature history must be established for the leads used.

Uncertainty in a difference includes covariance:This is the difference specialization of the primary uncertainty-propagation law.^[ref] With $u_t=u_0=1\mathrm{m\Omega }$ and correlation $0.9$, the difference uncertainty is about $0.447\mathrm{m\Omega }$; assuming independent readings would give $1.414\mathrm{m\Omega }$. Positive correlation can reduce a shared additive contribution, but a common gain error still scales the corrected specimen resistance.

An unchanged thermal offset also fails to cancel if the test currents differ: for $E_t=E_0=20\mu \mathrm{V}$, $I_t=10\mathrm{mA}$ and $I_0=1\mathrm{mA}$, its residual is $-0.018\Omega$. Current reversal within each arrangement can address a stable additive offset, but does not remove series resistance drift. Four-terminal sensing addresses the lead-drop geometry by measuring potential difference separately at the chosen specimen terminals.

A proposed check alternates short and specimen measurements while recording current and lead temperature, repeats after reconnecting the clips, and compares against a four-terminal result. The spread after reconnection tests a different contribution from rapid repeatability without disturbing the circuit. Report the defined specimen boundary, short construction and delay between readings rather than treating a successful zero display as a complete accuracy assessment.

A leave-one-out inverse-variance weighted consensus applies an inverse-variance weighted mean to all laboratories except the one being assessed. Suppose their results $X_j$ are independent random variables with common expected value $\theta$ and known positive variances $u_j^2$. For laboratory $i$, defineUnder this model it is the best linear unbiased estimator of $\theta$ from the remaining results. The omitted observation does not determine its own leave-one-out interlaboratory consensus value.

The minimum-variance property follows directly from the Cauchy-Schwarz inequality. For any fixed coefficients $c_j$ satisfying ${\sum }_{j\ne i}{c}_j=1$, the linear unbiased estimator $T={\sum }_{j\ne i}{c}_j{X}_j$ hasEquality holds precisely for $c_j=u_j^{-2}/W_{-i}$. ThereforeThis is a comparison among linear unbiased estimators under the stated independent-error model; it requires no normal distribution assumption.

For an original example, assess a laboratory reporting $x_i=12$ with standard uncertainty $u_i=0.5$. The other laboratories report $(10,12,8)$ with respective standard uncertainties $(1,2,4)$, all in the same units. Their total precision is $W_{-i}=1+1/4+1/16=21/16$, givingThe held-out difference is $D_i=12/7\simeq 1.714$, withThe denominator includes the benchmark's uncertainty even though $X_i$ was excluded. This uncertainty-normalized difference and a proficiency-test z-score based on a predetermined assessment scale answer different questions.

Shared bias in an interlaboratory consensus changes the uncertainty model. Suppose instead $X_j=\theta +B+{\epsilon }_j$, where the common reference contribution $B$ has variance $u_B^2$ and the mutually independent residuals, independent also of $B$, have variances $s_j^2$. Coefficients summing to one preserve $B$ in the benchmark. Its variance is $u_B^2+\sum c_j^2{s}_j^2$, so the minimizing coefficients use $s_j^{-2}$, andUsing the full marginal uncertainties $\sqrt{u_B^2+s_j^2}$ as if they were independent changes the coefficients and loses the shared-reference structure. The propagation of uncertainty retains the covariance of dependent inputs.^[ref]

If the held-out laboratory shares the same $B$, that contribution cancels in $D_i$, while it remains in the benchmark's absolute measurement uncertainty. Precision of a difference and knowledge of an absolute value are distinct properties. Estimated, result-dependent uncertainties or additional correlations require analysis of the actual estimator; the fixed known-variance proof does not establish optimality for those cases.

Logarithmic condition numbers of a tail-acquisition requirement quantify the fractional response of a positive recording extension to fractional parameter changes. For $L=\tau \ell$ with $\ell =\log (A\tau /\epsilon )>0$, define the signed sensitivity ${\kappa }_x=\partial \log L/\partial \log x$. ThenTheir absolute values are componentwise relative condition numbers. Acquisition-time sensitivity to tail-model parameters gives the corresponding dimensional derivatives and uncertainty propagation.

These numbers depend on which coordinates are held fixed. If $M=A\tau$ denotes the initial tail-energy allowance, then $L=\tau \log (M/\epsilon )$. In coordinates $(M,\tau ,\epsilon )$, the decay-time sensitivity at fixed $M$ is exactly one, rather than $1+1/\ell$. This is the same physical formula: holding $M$ fixed while changing $\tau$ also changes $A$. Transforming the parameter covariance consistently leaves the propagated output uncertainty unchanged.^[ref]

An original boundary example takes $A\tau /\epsilon =1.01$ and $\tau =20\mathrm{ns}$. The required extension is only $L\approx 0.1990\mathrm{ns}$, but ${\kappa }_A\approx 100.50$. Increasing $A$ by $1\%$ multiplies the ratio by $1.01$, giving $1.0201=(1.01{)}^2$ and doubling the required extension exactly. Its relative change is large because the original extension was very small.

This does not mean the absolute requirement is singular there. In the active branch $\partial L/\partial \log (A\tau /\epsilon )=\tau$ remains finite. At and below the threshold, the physical extension is zero, so $\log L$ is undefined. A logarithmic tail-duration change at the zero-extension boundary must instead use the piecewise requirement or absolute changes. Large relative conditioning alone is not evidence of a large absolute recording-time risk.

Finally, conditioning describes the stated envelope model, not its experimental validity. The optical power tail must be distinguished from detector response, which NIST measured independently in its pulse calibration procedure.^[ref] A well-conditioned calculation based on an invalid optical envelope still gives an unjustified acquisition guarantee.

Measurement calibration establishes a relation between an instrument's indications and reference values, with the relevant measurement uncertainty. It is distinct from changing the instrument until it reads correctly: an instrument can be calibrated by determining a response curve and applying a correction afterward.

A practical voltage experiment connects a meter to several reference values across its working range and repeats the sequence in both directions. Fitting $V_{\mathrm{indicated}}=a{V}_{\mathrm{reference}}+b$ estimates a scale factor and measurement zero offset; departures from that relation test the response model. Repetition at later times checks instrumental drift. Uncertainty in the reference, the fitted parameters, and unresolved departures belongs in the measurement uncertainty budget, even if the displayed readings appear very stable.^[ref]

A measurement model relates the measurand to measured inputs and correction quantities, for example $R=V/I$ when inferring electrical resistance. Its mathematical form states which physical effects are included in the result.

For small uncertainties, linearization gives $u_R^2\approx u_V^2/I^2+V^2{u}_I^2/I^4-2V\mathrm{Cov}(V,I)/I^3$. The covariance term shows why shared calibration information matters. A model must also account for effects such as instrumental drift when they materially influence the inference.^[ref]

A measurement standard realizes a specified quantity with an assigned value and associated measurement uncertainty, and is used as a reference. It can be a physical artifact, a measuring system or a suitable reference material. The value and its measurement uncertainty apply under stated conditions and at the relevant time; a nominal label alone is insufficient.^[ref]

For an original comparison, take a reference resistor with assigned electrical resistance $R_s=100.000\Omega$ and standard uncertainty $u(R_s)=0.001\Omega$. A comparison measures a working resistor's ratio $q=R_x/R_s=1.00020$, with independent $u(q)=0.00001$. The measurement model givesand first-order propagation of uncertainty givesThe reference's measurement uncertainty remains in the result even if the ratio could be read with arbitrarily many digits.^[ref]

A practical comparison records reference conditions, connecting arrangements and repeated readings while checking effects such as instrumental drift. A primary measurement standard has a particular basis for its assigned value; a working reference need not be primary to transfer useful metrological traceability. Distinguish this physical reference from a written technical specification, which is another use of the word standard.

Measurement uncertainty describes the spread of values reasonably attributable to a specified measurand, using the available information. It differs from measurement error, which is a difference from a reference value: the uncertainty can be quantified even when that difference is unknown. A useful result therefore states what was measured, the unit of measurement, the value, and how its uncertainty was evaluated.^[ref]

For a wire's electrical resistance obtained from $R=V/I$, uncertainty in the voltage and electric current propagates through that measurement model. Shared measurement calibration effects can make the inputs correlated, so adding percentages or treating every contribution as independent can be wrong. The measurement uncertainty budget records those assumptions. Report a standard uncertainty when a standard-deviation scale is intended, or an expanded uncertainty with its stated coverage factor; a bare plus-or-minus number does not specify which was used.

A record of uncertainty contributions in a specified measurement model, their evaluation methods, sensitivity coefficients and relevant correlations. For $y=g(x_1,\dots ,x_n)$, first-order propagation of uncertainty gives $u_y^2={\sum }_{ij}({\partial }_{i}g)({\partial }_{j}g)\mathrm{Cov}(x_i,x_j)$. A correction for a known systematic error can itself have uncertainty; applying the correction does not remove that contribution.^[ref]

Metrological traceability is a property of a measurement result: a documented, unbroken sequence of measurement calibrations relates it to a stated reference, and every link contributes to its measurement uncertainty. The reference can be a practical realization of a measurement unit, a measurement procedure or a measurement standard.^[ref]

Consider an original mass-measurement chain: a reference measurement standard calibrates a working mass, which is used for instrument calibration of a balance, which weighs an object. Suppose the reference contributes $2\mathrm{mg}$ of standard uncertainty, the transfer adds an independent $3\mathrm{mg}$, and the final weighing adds an independent $4\mathrm{mg}$. With unit sensitivity to these additive contributions,The intermediate working mass has $\sqrt{13}\mathrm{mg}$ of standard uncertainty. That value already includes the reference contribution; adding the same $2\mathrm{mg}$ again at the final step would double-count it. Correlated contributions require covariance terms rather than the independent-input calculation.^[ref]

Documentation must identify the reference, comparison procedures, dates, applicable conditions and uncertainty contributions. A certificate or instrument label alone does not supply the complete chain for every later result. Metrological traceability also does not ensure that the resulting measurement uncertainty is small enough for the intended use.

Minimum-noise weights under baseline moment constraints choose a linear pulse estimator with the smallest variance among those that preserve a specified signal and remove specified backgrounds. Write the reading vector as $\mathbf{x}=\theta \mathbf{s}+H\mathbf{\beta }+\mathbf{ϵ}$, where $\mathbf{s}$ is a known signal template, the columns of $H$ contain baseline shapes, and the error covariance matrix $C$ is positive definite. For $\hat{\theta }={\mathbf{a}}^T\mathbf{x}$, requireColumns $1,t,t^2,\dots$ in $H$ implement the moment conditions for baseline-drift cancellation. The noise cost of polynomial baseline cancellation depends on which feasible weights are chosen.

If the constraint columns are independent, Lagrange multipliers giveFor any feasible perturbation $\mathbf{v}$, $F^T\mathbf{v}=0$ and ${\mathbf{v}}^{T}C\mathbf{a}=0$, so the variance increases by ${\mathbf{v}}^{T}C\mathbf{v}$. This proves the minimum and connects the estimator to generalized least squares. The covariance propagation is the full quadratic form, not a sum that discards correlations.^[ref]

For an original one-bin pulse example, use five equally spaced times $-2h,-h,0,h,2h$, a signal present only at the centre, and independent errors of variance ${\sigma }^2$. To remove just constant and linear backgrounds, the optimal dimensionless weights are $(-1/4,-1/4,1,-1/4,-1/4)$. Their variance is $5{\sigma }^2/4$, lower than $3{\sigma }^2/2$ for the centre and its two nearest references alone. Multiplying the amplitude estimate by a bin duration $\Delta t$ gives energy variance $5\Delta t^2{\sigma }^2/4$. Extra uncontaminated references can reduce noise when the cancelled background family stays the same.

The inverse requires genuinely independent constraints. Constraint redundancy in baseline-noise minimization calls for removing redundant columns and checking feasibility; a signal template inside the nuisance span cannot be preserved while that span is rejected. NIST's before/after pulse baselines provide the physical measurement context.^[ref] The five-point design is an original idealization and assumes no pulse signal in its reference samples.

A minimum-variance spectral contrast orthogonal to a reference shape estimates a desired signal amplitude while exactly rejecting one fixed nuisance shape. Let $\mathbf{z}=\theta \mathbf{s}+\beta \mathbf{x}+\mathbf{ϵ}$, with known shapes $\mathbf{s},\mathbf{x}$ and a positive-definite noise covariance matrix $C$. Choose $\hat{\theta }={\mathbf{a}}^T\mathbf{z}$ to minimize ${\mathbf{a}}^{T}C\mathbf{a}$, subject toThe constraints retain unit signal response and reject the reference amplitude, as in projection of shared-scale uncertainty onto a spectral contrast. The quadratic variance includes the full noise covariance.^[ref]

Remove the reference component in the inverse-covariance geometry:For nonzero $\mathbf{x}$ and $D>0$, the minimizing contrast and its variance areFor a short proof, every other feasible contrast is $\mathbf{a}+\mathbf{v}$ with ${\mathbf{v}}^T\mathbf{s}={\mathbf{v}}^T\mathbf{x}=0$. Since $C\mathbf{a}$ is a linear combination of those shapes, the cross term vanishes, leaving the added nonnegative variance ${\mathbf{v}}^{T}C\mathbf{v}$. This is a constrained best linear unbiased estimator.

In an original three-channel example, take $\mathbf{s}=(1,0,0{)}^T$, $\mathbf{x}=(1,1,1{)}^T$ and $C=\mathrm{diag}(1,4,1)$. The optimum is $\mathbf{a}=(1,-1/5,-4/5{)}^T$, with variance $1.8$. Equal subtraction of the two reference channels, $(1,-1/2,-1/2{)}^T$, has variance $2.25$. The quieter third channel receives more of the subtraction.

If the desired and reference shapes are proportional, $D=0$ and no unit-response rejecting contrast exists. The noise penalty from nearly parallel signal and reference shapes already becomes large as $D$ approaches zero. This construction assumes fixed known shapes and covariance; estimating them from the same noisy spectra requires additional uncertainty terms.

The mismatch remaining after common RC gain cancellation is the timing error caused by unequal gains in the amplitude and threshold paths after their common multiplicative part has canceled. It depends on their ratio, so improving a shared gain alone does not remove it.

For the zero-baseline RC threshold-crossing time, let $a=\theta /H$ with $0<a<1$. If the amplitude is multiplied by $G_H>0$ and the threshold by $G_{\theta }>0$, define the residual relative error by $G_{\theta }/G_H=1+ϵ$. The exact result from ratiometric cancellation of RC timing gain errors isHere $\tau$ is the time constant. As $a(1+ϵ)$ approaches one, the crossing is arbitrarily late; at or above one there is no finite rising crossing. An error budget must check this existence condition as well as a small timing residual.

For a new example, set $H=1\mathrm{V}$, $\theta =0.8\mathrm{V}$ and $\tau =1\mu \mathrm{s}$. A one-percent positive mismatch gives $\Delta t=-\log (0.96)\mu \mathrm{s}\approx 40.822\mathrm{ns}$; a one-percent negative mismatch gives $-\log (1.04)\mu \mathrm{s}\approx -39.221\mathrm{ns}$. The first-order estimate is $\Delta t\simeq \tau aϵ/(1-a)$, or $\pm 40\mathrm{ns}$ here. The unequal exact magnitudes are a consequence of the logarithm, not an asymmetric tolerance specification.

If $|ϵ|\le e<(1-a)/a$, monotonicity gives the exact worst absolute errorThus a $10\mathrm{ns}$ bound in this example requires $e\le (1-a)(1-e^{-0.01})/a\approx 0.00248754$, about $0.249\%$. This is a bound on the gain ratio, not on either gain separately.

For small fractional path errors ${\eta }_H,{\eta }_{\theta }$, $ϵ\simeq {\eta }_{\theta }-{\eta }_H$ andA perfectly shared component has zero contribution to that difference. This is the law of propagation of uncertainty applied to the two gains; it does not cancel a change in $\tau$, an additive threshold error or a nonzero unscaled baseline.^[ref]

A proposed check varies the common gain while holding its ratio fixed, then varies only one path. Recording the two electric potential differences and the crossing on an oscilloscope distinguishes the exact common-gain invariance from the surviving mismatch.

Moment conditions for baseline-drift cancellation specify which polynomial backgrounds a linear measurement removes. Let readings at known times $t_i$ enter the corrected pulse estimate as $\hat{U}={\sum }_i{a}_i{x}_i$, with the integration/calibration units included in $a_i$. A background $b(t)={\sum }_{q=0}^p{c}_q{t}^q$ contributes nothing wheneverSubstitution gives this result directly. The zeroth condition removes a common offset; the first additionally removes linear drift. These conditions concern the actual weights and times, not only the number of background samples.

An original one-bin signal model measures a pulse at $t=0$, with uncontaminated reference samples at $-h$ and $h$. UseThe weights sum to zero and have zero first time moment. Constant and linear backgrounds cancel exactly, but a quadratic background $c_2{t}^2$ leaves bias $-\Delta t{c}_2{h}^2$. With chosen $h=2\mu \mathrm{s}$, $\Delta t=1\mu \mathrm{s}$ and $c_2=0.05mW/\mu {\mathrm{s}}^2$, the bias is $-0.20\mathrm{nJ}$. Here the pulse is stipulated to occupy the central bin; the formula is not a universal integral estimator for an arbitrary waveform.

Cancellation has a noise cost of polynomial baseline cancellation. For independent equal-variance reading errors ${\sigma }^2$, the chosen estimate has variance $(3/2)\Delta t^2{\sigma }^2$, even though its constant and linear bias terms vanish. General correlations use ${\mathbf{a}}^{T}C\mathbf{a}$ instead.^[ref]

NIST's historical pulse-energy apparatus used before/after baselines for background and thermal-drift compensation.^[ref] The polynomial test above is an original general analysis. Pulse contamination in a reference, uncertain timing or drift outside the assumed polynomial family can leave a residual even when the nominal algebraic conditions hold.

The negative covariance of empty and positive detector-interval counts arises because a fixed number of measurement intervals is divided into mutually exclusive categories. Let $E_i,P_i$ indicate that interval $i$ is respectively physically empty or strictly positive. The same interval cannot satisfy both, so $E_i{P}_i=0$ andFor $N$ independent repetitions with common probabilities, the counts $N_E={\sum }_i{E}_i$ and $N_P={\sum }_i{P}_i$ therefore haveThe joint probability of empty and strictly positive detector-width intervals gives the underlying multinomial model. The sign describes category-count dependence, not physical antagonism between two detectors.

For an original boundary example, take $N=100$ and $p_E=p_P=0.025$. Each count has variance $100(0.025)(0.975)=2.4375$, while their covariance is $-0.0625$. The total tail-category count has variance $2.4375+2.4375-0.125=4.75$. Treating the two counts as independent would give $4.875$ instead. The covariance is needed even though the individual measurement repetitions are independent.

Conditioning on the total is important. If the number of repetitions is random and independent of the category assignments, the law of total covariance givesAn independent total with a Poisson distribution has equal mean and variance, making this covariance zero; a more variable total can make it positive. This is a different sampling protocol, not a contradiction of the fixed-total formula.

Shared calibration drift is another change of model. Repetitions may then have dependent category indicators, and cross-repetition covariance terms must be added. Overdispersion of detector-interval counts under shared drift can make a simple multinomial analysis inadequate.

NIST's measurement guidance retains common influences in uncertainty propagation and requires assumptions for confidence-level interpretation.^[ref] State whether the repetition total is fixed, how the categories are defined and which calibration influences are shared before interpreting their empirical count covariance.

The noise cost of polynomial baseline cancellation is the change in random measurement uncertainty when extra background trends are removed. For a linear pulse estimate $\hat{U}={\mathbf{a}}^T\mathbf{x}$ and reading-error covariance matrix $C$,The moment conditions for baseline-drift cancellation constrain $\mathbf{a}$ to reject specified polynomial backgrounds. They do not make the remaining random error vanish.^[ref]

For an original comparison, assume a pulse occupies only the central sample and the other samples contain background alone. Three samples at $-h,0,h$ with weights $\Delta t(-1/2,1,-1/2)$ cancel constant and linear backgrounds. Five samples at $-2h,-h,0,h,2h$ can instead useThese weights preserve the central signal and have zero moments of orders $0,1,2,3$. For independent reading errors of equal variance ${\sigma }^2$, the three-sample variance is $3\Delta t^2{\sigma }^2/2$, while the five-sample variance isWith $\sigma =1\mathrm{mW}$ and $\Delta t=1\mu \mathrm{s}$, the standard uncertainties are approximately $1.225$ and $1.394\mathrm{nJ}$. The second design removes quadratic and cubic drift as well, but its larger reference weights increase noise. This compares two specified estimators; it is not a claim that collecting more background data must worsen precision.

NIST's pulse-energy apparatus used before/after baseline measurements for background and thermal-drift correction.^[ref] The polynomial comparison above is an original model for choosing such measurements. Choosing minimum-noise weights under baseline moment constraints can exploit additional reference samples and their measured correlations. Pulse leakage into a reference or drift outside the cancelled polynomial family still creates bias.

The noise penalty from nearly parallel signal and reference shapes is the precision lost when a spectral estimator must reject a nuisance pattern resembling its desired signal. For nonzero shapes $\mathbf{s},\mathbf{x}$ and positive-definite noise covariance $C$, define their inverse-covariance correlationThe Cauchy-Schwarz inequality gives $|r|\le 1$. The minimum-variance spectral contrast orthogonal to a reference shape has varianceIf the nuisance amplitude were known and subtracted without uncertainty, the corresponding optimum variance would be $V_{\mathrm{known}}=1/({\mathbf{s}}^T{C}^{-1}\mathbf{s})$. Exact nuisance rejection therefore multiplies variance by $1/(1-r^2)$ and standard uncertainty by $1/\sqrt{1-r^2}$.

The relevant angle is the noise-whitened angle between spectral signal and nuisance shapes: it is measured after a whitening transformation, not generally from the raw plotted spectra. A channel with large noise contributes little to separating the shapes even if their plotted curves differ there.

For an original example, let $C={\sigma }^{2}I$, $\mathbf{x}=(1,0{)}^T$ and $\mathbf{s}=(1,0.1{)}^T$. Rejecting the reference forces the contrast to $(0,10{)}^T$, with variance $100{\sigma }^2$. With known nuisance amplitude, its variance would be ${\sigma }^2/1.01$. The rejection penalty is therefore $101$ in variance, about $10.05$ in standard uncertainty. Only the weak second-channel component distinguishes signal from nuisance.

The primary stimulated-Raman experiment reported systematic spectral noise above its calculated shot-noise limit and described pure-solvent subtraction of a common pattern.^[ref] A reference can remove that shared pattern while also making an overlapping desired signal hard to estimate. Retain full covariance in the variance calculation.^[ref] At exact proportionality, no unit-response rejecting contrast exists; a low-noise result obtained by suppressing the signal as well is not a valid amplitude estimate.

Noise remaining after scaled Raman subtraction is the random uncertainty left when a surface contribution's mean has been cancelled. For two measured spectral channels $X,Y$ and fixed scale $k$, scaled subtraction of two spatially offset Raman spectra uses $Z=Y-kX$. Its exact variance isThe covariance term describes the actual acquisition relationship; numerical subtraction does not itself make the inputs correlated or independent. General measurement propagation retains such terms.^[ref]

A focused extension allows an uncertain subtraction scale $K$, independent of the pair $(X,Y)$, with mean $k$ and variance ${\sigma }_k^2$. Expanding the product's moments givesThe final term includes uncertainty in the reference scale, not just shot noise in the two spectra.

For an original surface-only example, let independent Poisson photoelectron counts have means ${\mu }_X=20000$ and ${\mu }_Y=6000$, with $k=0.30$. The mean residual is zero. A perfectly known scale gives variance $6000+0.30^2(20000)=7800$ and standard deviation $88.3$. If instead ${\sigma }_k=0.005$, the variance becomes $7800+0.005^2(20000^2+20000)=17800.5$, giving standard deviation $133.4$ in the scaled count units.

If $K$ is estimated from these same counts, the independence assumption needs replacement by the appropriate joint model. A scale shared across wavelengths can also produce correlated spectral residuals from a shared subtraction scale, leaving a structured band-like error rather than independent noise in each bin. The original SORS study demonstrates changing layer contributions with collection offset; the moment calculation here is an independent noise model for a simplified subtraction.^[ref]

The noise-whitened angle between spectral signal and nuisance shapes measures their separation in units of measurement noise rather than raw plotting units. For nonzero shapes $\mathbf{s},\mathbf{x}$ and a positive-definite covariance $C$, choose a square whitening transformation $W$ with $WC{W}^T=I$. The angle isIt is the ordinary Euclidean angle after whitening. Both parallel and antiparallel shapes are indistinguishable up to an unknown nuisance amplitude; the precision penalty depends on ${\sin }^2\varphi$, as in noise penalty from nearly parallel signal and reference shapes.

For an original example, take raw shapes $\mathbf{s}=(1,1{)}^T$, $\mathbf{x}=(1,-1{)}^T$ and $C=\mathrm{diag}(1,100)$. The plotted shapes are orthogonal, but whitening gives $(1,0.1{)}^T$ and $(1,-0.1{)}^T$. Thus $\cos \varphi =0.99/1.01\approx 0.98020$ and $\varphi \approx 11.42^{\circ }$. The second channel, which separates their signs, is noisy. Rejecting the nuisance multiplies the optimum variance by $1/{\sin }^2\varphi =25.5025$, or standard uncertainty by $5.05$.

The invariance of noise-whitened spectral angles to the whitening factor follows from $W^{T}W=C^{-1}$. Different valid square whitening factors can rotate the whitened coordinates without changing the angle. Likewise an invertible change of measurement units transforms both shapes and covariance, leaving the inverse-covariance expression unchanged. Rescaling only the displayed curves while leaving their noise model fixed is a different operation.

The covariance must describe the actual measured channels and shared errors, using the full propagation rule.^[ref] If $C$ is singular, first restrict to its physically allowed range; an unrestricted inverse whitening factor does not exist.

The primary stimulated-Raman experiment described systematic spectral noise and subtraction of a pure-solvent reference.^[ref] The angle above assesses which desired patterns remain distinguishable after such nuisance rejection under a stated covariance. It does not establish that an estimated reference shape or noise model is itself accurate.